JP2000134538A - Device and method for image processing and providing medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To extract a prescribed image area without deteriorating image activity. SOLUTION: A frame memory 1 stores a pixel value (YUV signal) for one frame of a supplied original image and supplies the pixel value of a pixel corresponding to a request from a pixel value comparator circuit 2 to the circuit 2. The circuit 2 respectively compares the values of the Y signal, U signal and V signal of adjacent pixels based on label information (coordinates and label) stored in a label memory 4 and outputs their results to a label setting circuit 3. The circuit 3 sets a label to a set (area coordinates) of adjacent pixels whose pixel values are equal based on the comparison results and outputs the label information to the memory 4. A pixel value averaging circuit 6 generates an upper hierarchical image of an image stored in the memory 1 through the control of a decision circuit 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image processing apparatus and method, and a providing medium, and more particularly to an image processing apparatus and method for extracting a predetermined area in an original image, and a providing medium.

[0002]

2. Description of the Related Art Conventionally, when extracting an area of adjacent pixels having substantially equal pixel values in an original image, a low-pass filter process is applied to the original image, or the original image is subjected to a predetermined process. The image is divided into pixel blocks of a large size, the pixel values are averaged for each block, and the resolution (activity) of the image is reduced, so that the boundary between the regions is ambiguous and the extraction processing is performed.

[0003]

Therefore, when a predetermined image area is extracted by using the conventional method, there is a problem that the boundary of the extracted area is ambiguous and accurate area division cannot be performed.

Further, there is a problem that the activity of the extracted image is deteriorated.

[0005] The present invention has been made in view of such a situation, and it is an object of the present invention to extract a predetermined image area without deteriorating the activity of an image.

[0006]

According to the first aspect of the present invention, there is provided an image processing apparatus comprising: a storage unit for storing an input image; and a comparison unit for comparing pixel values of adjacent pixels in the image stored by the storage unit. Means, and setting means for setting a label in an area where adjacent pixels are located corresponding to the comparison result of the comparing means,
A storage unit for storing the information of the label set by the setting unit, and an upper layer image of the image stored by the storage unit,
The storage unit includes an updating unit that updates the image stored so far, and a control unit that controls the updating unit based on label information stored by the storage unit.

According to a fourth aspect of the present invention, there is provided an image processing method comprising: a storing step of storing an input image; a comparing step of comparing pixel values of adjacent pixels in the image stored in the storing step; A setting step of setting a label in an area where adjacent pixels are located in accordance with the comparison result in step (a), a storage step of storing information of the label set in the setting step, and an upper layer image of the image stored in the storage step. An update step of generating and updating an image stored so far in the storage step, and a control step of controlling generation of an upper layer image in the update step based on information of a label stored in the storage step. Features.

According to a fifth aspect of the present invention, there is provided a providing medium, comprising: a storing step of storing an input image; a comparing step of comparing pixel values of adjacent pixels in the image stored in the storing step; A setting step of setting a label in an area where adjacent pixels are located in accordance with the comparison result, a storage step of storing information of the label set in the setting step, and an upper layer image of the image stored in the storage step An update step of updating the image stored so far in the storage step, and a control step of controlling generation of an upper layer image in the update step based on the information of the label stored in the storage step. A computer-readable program to be executed by a processing device is provided.

An image processing apparatus according to claim 1, wherein
In the image processing method described in the above, the input image is stored, and the pixel values of adjacent pixels in the stored image are compared to correspond to the comparison result. A label is set in an area where adjacent pixels are located, and information of the set label is stored. Further, an upper layer image of the stored image is generated, the image stored so far is updated, and the generation of the upper layer image is controlled based on the stored label information.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below. In order to clarify the correspondence between each means of the invention described in the claims and the following embodiments, each means is described. When the features of the present invention are described by adding the corresponding embodiment (however, an example) in parentheses after the parentheses, the result is as follows. However, of course, this description does not mean that each means is limited to those described.

An image processing apparatus according to a first aspect of the present invention includes a storage unit (eg, the frame memory 1 in FIG. 1) for storing an input image, and a pixel value of an adjacent pixel in the image stored by the storage unit. (For example, the pixel value comparing circuit 2 in FIG. 1 and setting means for setting a label in an area where adjacent pixels are located corresponding to the comparison result of the comparing means (for example, the label setting circuit in FIG. 1) 3), storage means (for example, the label memory 4 in FIG. 1) for storing the information of the label set by the setting means, and an upper hierarchical image of the image stored in the storage means, which is stored by the storage means until then. Updating means (for example, the pixel value averaging circuit 6 in FIG. 1) for updating the stored image and information on a label stored by the storing means,
And a control unit (for example, the determination circuit 5 in FIG. 1) for controlling the update unit.

An example of the configuration of an image extracting apparatus to which the present invention is applied will be described with reference to FIG. The original image data (YUV signal) input to the image extraction device is supplied to the frame memory 1 and the image extraction circuit 8. The frame memory 1 stores a pixel value (YUV signal) for one frame of the supplied original image, and supplies the pixel value of a pixel corresponding to a request from the pixel value comparison circuit 2 to the pixel value comparison circuit 2.
Further, the frame memory 1 supplies the pixel value of the pixel corresponding to the request from the pixel value averaging circuit 6 to the pixel value averaging circuit 6. Further, the frame memory 1 uses the pixel value (described later) of the upper layer image input from the pixel value averaging circuit 6 to
The pixel value of the corresponding pixel stored so far is updated.

The pixel value comparison circuit 2 compares the values of the Y signal, U signal, and V signal of adjacent pixels based on label information (described later) stored in the label memory 4, and compares the comparison results. Is output to the label setting circuit 3. The label setting circuit 3 has the same pixel value based on the comparison result of each signal component of the adjacent pixel value input from the pixel value comparison circuit 2 and generates a set of adjacent pixels (the coordinates of the area). Then, a label is set, and the label information (coordinates and label) is output to the label memory 4.

The label memory 4 stores (stores) a corresponding label at a memory address corresponding to the coordinates of the label information input from the label setting circuit 3.

The determination circuit 5 refers to the label information stored in the label memory 4 to determine whether or not there is a coordinate with no label set. If determined, the pixel value averaging circuit 6
To generate an upper layer image of the image stored in the frame memory 1. When the determination circuit 5 determines that there is no coordinate for which a label is not set (a label is set for all coordinates), the switch 7
Is turned on, the label information stored in the label memory 4 is output to the image extraction circuit 8.

The pixel value averaging circuit 6 generates an upper layer image of the image stored in the frame memory 1 under the control of the determination circuit 5. That is, the pixel value averaging circuit 6 reads out the pixel values of the image stored in the frame memory 1 and averages the plurality of pixel values for each predetermined block (for example, 2 × 2 pixels). The average value (the pixel value of the upper hierarchical image) is output to the frame memory 1 as the pixel values of a plurality of pixels constituting the block.

An image extracting circuit 8 extracts a pixel value of a pixel of an original image corresponding to a coordinate set with the same label based on the label information input via the switch 7 and outputs the pixel value to a subsequent stage. I do.

Next, the area identifying process of the image extracting apparatus will be described with reference to the flowchart of FIG. In step S1, the pixel value (YUV signal) of one frame of the original image input to the image extraction device is supplied to the frame memory 1 and the image extraction circuit 8. The frame memory 1 and the image extraction circuit 8 store the input pixel values of one frame of the original image.

In step S2, the pixel value comparison circuit 2
Reads out pixel values of left and right adjacent pixels from the frame memory 1 for each horizontal line of an image, compares each signal component (Y signal, U signal, and V signal) with each other. Are output to the label setting circuit 3 as shown in FIG.

In step S3, the label setting circuit 3
Is based on the amount of change of each signal component input from the pixel value comparison circuit 2, whether each signal component of two or more pixels consecutively adjacent in the horizontal direction is equal to each other (total signals of two or more adjacent pixels). Is determined to be 0) or not.
If it is determined that the signal components of two or more adjacent pixels are equal, the process proceeds to step S4.

In step S4, the label setting circuit 3
Sets a serial label ian in order from the left side of the line for each set of horizontally adjacent pixels (linear regions) having the same pixel value, and outputs the label information to the label memory 4. Note that i of the label ian indicates a vertical position of the line, and n indicates a serial number for each line.

In step S5, the pixel value comparison circuit 2
Determines whether or not the processing of steps S2 to S4 described above has been performed on all the lines of the image stored in the frame memory 1 and has performed those processings on all the lines. Until the determination, steps S2 to S4
Is repeated. If it is determined in step S5 that those processes have been executed, the process proceeds to step S6.

In step S6, the pixel value comparison circuit 2
Reads the label information of a predetermined line and the line below it from the label memory 4 and, based on the label information, the areas 1a0 and 2a0, the areas 1a1 and 2a shown in FIG.
1. The pixel values of two areas having vertically adjacent labels, such as the areas 1a2 and 2a2 and the areas 1a4 and 2a3, are read from the frame memory 1 and compared for each signal component. This comparison result is output to the label setting circuit 3.

In step S7, the label setting circuit 3
Refers to the comparison result of the pixel values of the pixel values of the vertically adjacent regions input from the pixel value comparison circuit 2 to determine whether the signal components of the pixel values of the vertically adjacent regions are equal to each other. If it is determined that the signal components of the pixel values in the vertically adjacent area are equal to each other, in step S8, the label of the area of the lower line is updated using the label of the area of the upper line, and the update is performed. The label information thus output is output to the label memory 4.

For example, the region 1a0 and the region 2a shown in FIG.
0, and when the respective signal components of the pixel values of the area 1a2 and the area 2a2 are equal to each other, the label of the area 2a0 is updated to 1a0, and the label of the area 2a2 is updated to 1a2, as shown in FIG. You.

In step S9, the pixel value comparison circuit 2
Determines whether or not the above-described processing of steps S6 to S8 has been performed on all the lines of the image stored in the frame memory 1, and has determined that the processing has been performed on all the lines. Until the determination, steps S6 to S8
Is repeated. If it is determined in step S9 that the processing has been performed on all the lines,
Proceed to step S10.

In step S10, the judgment circuit 5
Referring to the label information stored in the label memory 4,
It is determined whether or not there are coordinates for which no label is set. If it is determined that there are coordinates for which no label is set, the process proceeds to step S11.

In step S11, the pixel value averaging circuit 6 reads out the pixel values of the image stored in the frame memory 1 under the control of the determination circuit 5, and reads out the pixel values for each predetermined block (for example, 2 × 2 pixels). A plurality of pixel values are averaged, and the obtained average value is output to the frame memory 1 as pixel values of a plurality of pixels constituting the block. The frame memory 1 updates the stored pixel value using the input pixel value. Therefore, the frame memory 1 stores the pixel values of the higher hierarchical image (coarse image) in which the pixel values of the pixels in the pixel block are equal, instead of the pixel values of the original image stored up to that point.

In step S12, the pixel value comparison circuit 2 reads the label information of a predetermined line from the label memory 4, and sandwiches the label information between the areas ia3 and ia4 as shown in FIG. The pixel value of the area ia3 and the pixel value of the pixel in contact with the area ia3 of the unlabeled area are read from the frame memory 1, and the respective signal components of the two are compared. I do. Similarly, the pixel value comparison circuit 2 reads out the pixel value of the area ia4 and the pixel value of the pixel in contact with the area ia4 of the unlabeled area from the frame memory 1, and compares the two signal components. The comparison result is output to the label setting circuit 3.

In step S13, the label setting circuit 3 refers to the comparison result of the pixel value at the boundary of the area where the label is not set, which is input from the pixel value comparison circuit 2, and determines the area where the label is not set. It is determined whether the pixel value at the end of is equal to each signal component of the pixel value of the adjacent, label-set area, and if it is determined that both signal components are equal, the process proceeds to step S14. , Set the same label as the area where the adjacent label is set to the coordinates of the pixel at the end of the area without the label (the coordinates of the pixel at the end of the area without the label and the area where the adjacent label is set) Included).

For example, when the pixel value of the area ia3 shown in FIG. 5A is equal to the pixel value of the pixel in contact with the area ia3 of the unlabeled area, there is no label as shown in FIG. 5B. The label ia3 is set at the coordinates of the leftmost pixel of the region (the coordinates of the leftmost pixel of the region without the label are included in the region ia3).

Thereafter, the flow returns to step S1, and the subsequent processing is executed. The processing in steps S1 to S14 is repeated until it is determined in step S10 that there is no coordinate for which no label is set. If it is determined in step S10 that there is no coordinate for which a label is not set, in step S15, the determination circuit 5 turns on the switch 7 and sets the label memory 4
Causes the image extraction circuit 8 to output the label information stored therein.

If it is determined in step S3 that any of the signal components of the adjacent pixels is not equal, the processing in step S4 is skipped. If it is determined in step S7 that any of the signal components of the pixel values of the vertically adjacent regions is not equal, the process of step S8 is skipped. Further, when it is determined in step S13 that any of the signal components of the pixel values at the boundary of the region where no label is set is not equal, the processing in step S14 is skipped.

Note that the above steps S3, S7, S1
In 3, it is determined whether the signal components of the adjacent pixels are equal to each other. However, it is determined whether the difference between the signal components of the adjacent pixels is smaller than a predetermined minute reference value. May be determined.

If the image extracting apparatus of the present embodiment is used instead of the conventional so-called chroma key processing, it is possible to easily perform image synthesis.

The present invention can be applied to a device for detecting a motion vector of an object by extracting the object from a continuous frame image.

Further, when the present invention is applied to, for example, a videophone, only a human image can be extracted. If only the extracted image is transmitted, the amount of information to be transmitted can be reduced. Becomes

It should be noted that the computer program for performing each of the above-described processes can be provided to the user via a network medium such as the Internet or a digital satellite, in addition to a medium such as a magnetic disk or a CD-ROM. it can.

[0039]

As described above, according to the image processing apparatus according to the first aspect, the image processing method according to the fourth aspect, and the providing medium according to the fifth aspect, the pixel values of the adjacent pixels are determined. Compare and set a label in the area where the adjacent pixels are located according to the comparison result, and based on the information of the set label, generate the upper layer image, so that the activity of the image is not degraded , A predetermined image area can be extracted.

[Brief description of the drawings]

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

FIG. 2 is a flowchart illustrating an area identification process of the image extraction device in FIG. 1;

FIG. 3 is a diagram illustrating a process of a label setting circuit 3 of FIG. 1;

FIG. 4 is a diagram illustrating a process of a label setting circuit 3 of FIG. 1;

FIG. 5 is a diagram illustrating a process of a label setting circuit 3 in FIG. 1;

[Explanation of symbols]

1 frame memory, 2 pixel value comparison circuit, 3 label setting circuit, 4 label memory, 5 judgment circuit,
6 pixel value averaging circuit, 7 switch, 8 image extraction circuit

Continued on the front page F-term (reference)

Claims (5)

[Claims] In an image processing apparatus for extracting a predetermined area in an original image, a storage unit for storing an input image and a pixel value of an adjacent pixel in the image stored by the storage unit are compared. Setting means for setting a label in an area where the adjacent pixel is located in accordance with the comparison result of the comparing means; storage means for storing information of the label set by the setting means; Generating an upper layer image of the image stored in the storage means,
Image processing, comprising: updating means for updating the image stored by the storage means until then; and control means for controlling the updating means based on information of the label stored by the storage means. apparatus. 2. The control unit controls the setting unit to set labels for all areas of the image based on information of the label stored by the storage unit. The image processing device according to claim 1. 3. The image processing apparatus according to claim 1, wherein the setting unit updates the information of the label stored by the storage unit in accordance with a comparison result of the comparison unit. 4. An image processing method for extracting a predetermined area in an original image, comprising: a storage step of storing an input image; and a pixel value of an adjacent pixel in the image stored in the storage step. A comparing step of comparing, a setting step of setting a label in an area where the adjacent pixels are located corresponding to the comparison result in the comparing step, and a storing step of storing information of the label set in the setting step. An update step of generating an upper layer image of the image stored in the storage step and updating the image stored so far in the storage step, based on information of the label stored in the storage step, A control step of controlling generation of the upper layer image in the updating step. 5. An image processing apparatus for extracting a predetermined area in an original image, comprising: a storage step of storing an input image; and a pixel value of an adjacent pixel in the image stored in the storage step. A comparing step of comparing, a setting step of setting a label in an area where the adjacent pixels are located corresponding to the comparison result in the comparing step, and a storing step of storing information of the label set in the setting step. An update step of generating an upper layer image of the image stored in the storage step and updating the image stored so far in the storage step, based on information of the label stored in the storage step, And a control step of controlling the generation of the upper layer image in the updating step. A providing medium characterized by providing a ram.