KR0137923B1 - Changed region abstraction method using adjacent pixel - Google Patents

Abstract

The present invention divides a current image into a region having a similar value in order to extract a changed region generated by the movement of an object, and obtains a difference between the images using only the surrounding pixels of the region. The present invention relates to a method of extracting a motion region using surrounding pixels for determining whether the image is detected. Since the entire area of the current image is divided into one region, the number of pixels determined to be a moving region is smaller than the number of pixels determined to be an unmoved region. In the present invention, a problem that is determined as a non-moving region occurs even in the fine movement. In the present invention, the current image is divided into regions having similar values, and the movement region extracted by the difference between the current image and the previous image has a similar value. The motion region is determined by using the information divided by regions. It is necessary to extract the motion region such as video phone, video conferencing system video encoder and decoder by judging whether it is a motion region by using only the surrounding pixels in the region divided into similar values in the current image. In this system, the image quality is improved by effectively extracting the motion area, which increases the reliability of the product.

Description

Motion region extraction method using surrounding pixels

1 is a block diagram of a general motion region extraction apparatus,

Second degree of movement of the object,

3 is a state diagram of a difference signal between images;

4 is a pixel used as a criterion in the conventional method,

FIG. 5 is a pixel determined as a motion area in FIG. 4;

6 is a block diagram of a motion region extraction apparatus according to the present invention.

7 is a pixel used as a criterion in the present invention;

FIG. 8 illustrates pixels determined as a motion area in FIG. 7.

9 is a diagram showing a morphological filter for explaining the present invention.

＊ Description of symbols for main parts of the drawings ＊

1: frame difference extraction unit 2: threshold processing unit

3: middle value filter part 4: small area removal part

5: Image segmentation unit 6: Motion region determination unit for each region

7: surrounding area detection unit

According to the present invention, a segment obtained by segmenting a current image into a homogenous region in order to extract a changed region generated by an object moving, and a value obtained by obtaining a difference between the images may be obtained. The present invention relates to a method for extracting a motion region using surrounding pixels to determine whether a moving region is used using only surrounding pixels.

Here, the motion area means an image changed through movement between the current image and the previous image.

A general motion image extracting apparatus is constructed as shown in FIG. 1, which is proposed in the papers of HOETER and DIEHL.

That is, the frame difference extracting unit 1 which obtains the difference between the current image and the previous image, and the window obtained with respect to the value obtained by the frame difference extracting unit 1 are formed around the corresponding pixel to average the absolute value in the window. If the threshold value is greater than or equal to the threshold value processing unit 2 that determines the center pixel as the movement region, and the intermediate value filter unit 3 which creates a window around the pixel and replaces the center value of the window with the intermediate value in the window. ), A region removal unit 4 for removing a small isolated region among regions determined to be a movement region and a region determined to not move, and a current value input to the frame difference extracting unit 1 are similar values. The image segmentation unit 5 and the region removing unit 4 for dividing the image into regions having a motion region determine whether the region is a motion region for each region obtained by the image divider 5. Consists of a decision unit (6).

The operation of the motion region extraction apparatus configured as described above is as follows.

First, the difference between each image is obtained by inputting the previous image and the current image into the frame difference extracting unit 1, and then the threshold value processing unit 2 creates a window centering on the corresponding pixel for the difference of the absolute image in the window. If the average value is greater than or equal to the threshold value, a process of determining the center pixel as the moving area is performed, and the median filter unit 3 creates a window centered on the pixel and replaces the median value with the median value in the window. Then, the area removing unit 4 extracts the moving area by removing an isolated small area among the area determined as the moving area and the area determined as the non-moving area.

The image splitter 5 divides the current image input to the frame difference extractor 1 into regions having similar values, and moves the motion image extracted by the region remover 4 by the divided regions. Determine if it is an area.

In determining whether it is a moving area as described above, the object moved as shown in FIG. 2 obtains a third degree when the difference between the current image and the previous image is obtained by the frame difference extracting unit 1. Although the center part of FIG. 3 is a moving area, the values of the current image and the previous image are the same. When the image segmentation unit 5 of the apparatus for extracting a motion region of FIG. 1 divides the current image into regions having similar values, the entire region of the current image is divided into one region as shown in FIG. 4. In this case, if the number of pixels in the corresponding area is larger than the existing number of pixels using the entire pixels belonging to the entire area, the existing method is regarded as the moving area. Since the number of pixels is smaller than the number of pixels, a problem arises in that the region of FIG. 4 is determined as the non-moving region.

In view of this, the present invention divides the current image into regions having similar values, and uses the information obtained by dividing the motion region extracted by the difference between the current image and the previous image into regions having similar values. In the method of determining recognition, a method of selecting a pixel to be used as a criterion for determining a motion region is characterized.

That is, in the method of determining whether the current image is divided into regions having similar values, and whether the motion region extracted by the difference between the current image and the previous image is divided into regions having similar values, and determining whether the motion region is a motion region. The method determines whether the corresponding area is a moving area or an unmoving area by using only the surrounding pixels of the area in which the current image is divided.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 6 is a block diagram of the present invention. In the existing configuration of FIG. 1, a peripheral region detector 7 is further provided between the image divider 5 and the region-specific motion region determiner 6. The same reference numerals are used, and description thereof will be omitted.

First, when the image segmentation unit 5 of the apparatus for extracting a motion region of FIG. 1 divides the current image into regions having similar values, the entire region of the current image is divided into one region as shown in FIG. 4. In the region-specific motion region determiner 6 of FIG. 6, the motion region is determined by using the inter-frame changed region which is the output of the small region remover 4 and the surrounding region detected by the surrounding region detector 7.

If the number of pixels changed between frames located in the surrounding area is larger than the reference number of pixels in the surrounding area, the corresponding area is determined to be a moving area.

The corresponding region refers to a divided region corresponding to the peripheral region used here among the regions divided by the image divider 5.

Here, FIG. 7 shows only peripheral pixels in the region of the current image, and FIG. 8 shows a portion of the boundary pixels, which is determined as a movement region, and a pixel region, which is determined as an unmoved region.

Considering only the boundary pixels in FIG. 8, the entire region becomes the movement region because the number of elements that are determined as the moving region is greater than that of the non-moving region.

Accordingly, even if there are many overlapping parts between the previous image and the current image, that is, even the fine movement, the movement region can be accurately extracted.

That is, in the present invention, the method for detecting the surrounding area using the surrounding area detecting unit 7 may be generally used, and the present invention may use a morphological filter as shown in FIG. 9. For the image segmented through 6-degree image segmentation (5), use the same filter as in FIG. 9 to place the center of the filter on each pixel of the image, and 1 of the 8 pixels around when the pixel located in the center is within the segmented image. If there are pixels outside the region, even one pixel is selected, the pixel in the center of the filter is selected as the peripheral region pixel.

As described above, according to the present invention, the image quality is improved by effectively extracting the motion region in a system that needs to extract the motion region, such as a video telephone, a video conferencing system video encoder, and a decoder, thereby increasing the reliability of the product.

Claims (2)

In the image divided by the image splitter 5, the motion region determination unit 6 for each region is performed by using the peripheral region detected by the surrounding image detector 7 and the changed region between frames, which is the output of the small region remover 4. ), When the number of pixels changed between frames located in the surrounding area is larger than the reference number of pixels compared to the number of pixels in the surrounding area, the corresponding area is determined as the moving area. . In selecting a peripheral region pixel using a filter for an image passing through the image splitter 5, 1 is selected from among 8 pixels when the center of the filter is placed in each pixel of the image and the pixel located at the center is in the divided region. A method of extracting a motion region using surrounding pixels, wherein a pixel at the center of the filter is selected as the surrounding region pixel when there are pixels outside the region even if there are four pixels.