JPH02118884A - Separating and partitioning device for moving area - Google Patents

Abstract

PURPOSE:To separate and partition the internal part of a moving area by providing a moving vector detecting circuit, a direction arithmetic circuit, and a moving area uniting circuit and outputting respective separated moving areas as label pictures to which respective different numbers are attached. CONSTITUTION:A moving vector detecting circuit 3 detects the moving vector between frame memories 1 and 2 for a picture, the direction component of the detected moving vector is calculated by a direction arithmetic circuit 4, and the direction component of the moving vector is outputted to a moving area uniting circuit 5. In the circuit 5, a threshold is provided, and picture elements are partitioned into areas by deciding that two adjoining picture elements are in the same area when the difference between the direction elements of the two adjoining picture elements is <= threshold and deciding that the two adjoining picture elements are in respectively different areas when the above-mentioned difference is > the threshold. Further, the calculated results are outputted as the label pictures in which the same area is displayed with the same number and stored in a memory 6 for output. Thus, the internal part of the moving area can be separated and partitioned, and processing can be executed in units of the respective areas by uniting the area with the differences between the moving vectors of the respective picture elements as the objects.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a moving region separation and division device used in a moving image analysis device, a moving image encoding device, and the like.

BACKGROUND OF THE INVENTION Conventionally, this D-heel motion region separation and division device has two image frame memories, a motion vector detection circuit, a static motion separation circuit,
It is equipped with a frame memory for output images, and is configured to be able to separate a still area and a moving area depending on the presence or absence of a motion vector.

FIG. 4 shows the configuration of a conventional motion region separation/division device. In Fig. 4, 11.12 is an image frame memory;
13 is a motion vector detection circuit, 14 is a static motion separation circuit, 1
5 is an output memory. Next, the operation of the above conventional example will be explained. The motion vector detection circuit 13 compares the contents of the image frame memo January 1 with the contents of the image frame memo 1) 12, and detects the movement between images as a motion vector. A motion vector is the amount of movement in pixel units, and the detected motion vector is sent to 14 static-motion separation circuits.

The static-motion separation circuit compares the magnitude of the motion vector with a threshold, and if the magnitude of the motion vector is smaller than the threshold, it is determined to be a stationary area, and conversely, if it is larger than the threshold, it is determined to be a moving area, and the result is divided into 15 data is transmitted to the output memory of

In this way, even with the above-mentioned conventional motion region separation and division device, it is possible to separate the static region and the motion region.

Problems to be Solved by the Invention However, the above-mentioned conventional motion region separation and division device only separates and divides a still region and a motion region based on the presence or absence of a motion vector, and only separates and divides the inside of a motion region by considering the similarity of motion. The problem was that I couldn't do it.

The present invention solves these conventional problems and aims to provide an excellent motion region separation and division device that can separate and divide the inside of a motion region.

Means for Solving the Problems In order to achieve the above object, the present invention provides a motion vector direction calculation circuit and a motion region merging circuit, and outputs the separated motion regions as labeled images with different numbers. It was designed to do so.

Therefore, according to the present invention, for the direction component of a motion vector, a motion region can be separated and divided using the angular difference with neighboring components as a threshold, and the motion region separation and division results that take into account the similarity of motion can be divided into This has the advantage that each region can be obtained as a labeled image with a number attached to it.

Embodiment FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In Fig. 1, 1 and 2 are image frame memories, 3
4 is a motion vector detection circuit, 4 is a direction calculation circuit, 5 is a motion area merging circuit, and 6 is an output memory.

Next, the above embodiment will be explained. A motion vector detection circuit 3 detects a motion vector between the image frame memories 1 and 2. The direction component θ of the detected motion vector is calculated by the direction calculation circuit 4. FIG. 2 shows an example of a direction calculation circuit. The motion vector may be given as an X-direction component and a Y-direction component of the motion vector, or it may be a predetermined representative number of the motion vector.

The output θ of the direction calculation circuit is sent to the motion area merging circuit 5. In the dynamic region merging circuit 5, the threshold value is set to θt
As b, if the difference in direction components between adjacent pixels is less than or equal to a threshold value, the pixels are divided into the same region, and if the difference exceeds the threshold value, the pixels are divided into different regions.

Figure 3 shows an overview of the region merging algorithm. The result of the calculation as described above is output as a label image in which the same area is displayed with the same number, and is stored in the output memory 5.

Effects of the Invention As is clear from the above embodiments, the present invention has the following effects.

Since the regions are merged with the difference in the directional components of the motion vectors of each pixel being symmetrical, it is possible to separate and divide the inside of the moving region, and it is possible to perform processing on a unit-by-region basis.

[Brief explanation of the drawing]

FIG. 1 is a schematic block diagram of a moving image separation and division device according to an embodiment of the present invention, FIG. 2 is a block diagram of a direction calculation circuit of the same device, FIG. 3 is a diagram explaining a region merging algorithm, and FIG. 4 1 is a schematic block diagram of a conventional moving image separation and division device. ■, 2 - Frame memory for images, 3... Motion vector detection circuit, 4... Direction calculation circuit, - Motion area merging circuit, 6... Output memory. Name of agent: Patent attorney Shigetaka Awano and 1 other person

Claims (1)

[Claims] a motion vector detection circuit that detects displacement of corresponding pixels between adjacent image frames constituting a moving image, a direction calculation circuit that calculates the displacement direction of the pixel based on the motion vector, and a direction calculation circuit that calculates the displacement direction of the adjacent pixel. The difference is compared with a certain threshold, and when the difference in the displacement direction is larger than the threshold, it is determined that they are different regions, and when it is smaller than the threshold, it is determined that they are the same region, and the same regions are successively merged to create a video. A moving region separating and dividing device comprising a moving region merging circuit that divides an image into a plurality of regions having the same displacement direction.