JPS60232787A - Picture synthesizing device - Google Patents

Abstract

PURPOSE:To form easily a synthesized picture with high quality by synthesizing the contents of a picture memory in accordance with an output signal from a picture border processor. CONSTITUTION:A picture memory 1 accumulates pictures from a picture input device 9 such as a TV camera or the like. A key signal memory 2 holds a key signal for extracting only an objective picture when plural pictures are synthesized, the key signal is set up to a binary value or a multi-level value in accordance with the shape of the objective picture and the shape of the objective picture can be inputted from a key signal input circuit 7. Receiving the output of the memory 2, the picture border processor 3 expresses the border part smoothly and supplies its output to a picture generator 4 together with the output of the memory 1. The picture generator 4 extracts or mixes the corresponding plural pictures in the memory 1 in accordance with the key signal modulated by the processor 3 to synthesize a picture and display the synthesized picture on a picture display 10.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention is an image that generates a key signal for extracting a target image and uses it to eliminate unnaturalness in the boundary area and create a high-quality composite image. This relates to a synthesis device.

Conventional configuration and its problems In recent years, hardware related to image processing devices.

As the functionality of software has greatly improved, the number of application fields that require high quality output images is also increasing. When trying to improve the quality of images, a point that requires special attention is how to perform processing to make unnatural displays such as ``jaggies'' in information such as bit maps and displays less noticeable. .

Conventionally, there are the following methods, for example. 1. First, a target image is captured from a television camera into an image memory with a blue background. Next, the red and green of that image.

Substitute the density of blue (represented as R, G, B) into a certain judgment formula, compare the value with a specified threshold, identify whether it is the target image or the background, and adjust the mixing ratio around the boundary. and combine it with other images. For example, it is possible to blur the image by changing the mixing ratio linearly around the boundary.

However, in the above method, since the entire screen is directly processed using image data, the @ area of the target image and the background area change with a very small difference due to the decision algebraic threshold, and the stitch pattern In such delicate cases, there is a high possibility that the image itself will become unclear, causing the problem that fine information contained in the image will be lost.

OBJECTS OF THE INVENTION It is an object of the present invention to provide an image compositing device that generates a natural, high-quality composite image with few jaggies in the boundary portions without significantly impairing detailed image information.

Structure of the Invention The image synthesis device of the present invention includes an image memory for storing a plurality of images, a key signal memory for storing a key signal which is a control signal for extracting a target image from an image signal, and the contents of the key signal memory. an image boundary processing device that generates a control signal for smoothing the boundaries of a composite image; and an image generation device that synthesizes a plurality of images from the contents of the image memory and the output of the image boundary processing device; It is configured to generate a composite image in which the boundary portions do not look unnatural, and it is possible to obtain an image with less jaggies without losing the information contained in the image.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings.

The drawing shows a block diagram of main parts of an image synthesis device according to an embodiment of the present invention. In the drawings, reference numeral 1 denotes an image memory that stores images input from an image input device 9 such as a television camera, a tablet, or other image files. A 2-digit key signal memory stores key signals for extracting only the target image when composing multiple images.

The key signal is set to be binary or multivalued depending on the shape of the target image. The shape of the target image can be input by the user through the key signal input circuit 7, or can be input in advance by another method, for example, by a contour extraction device. Further, the key signal can be arbitrarily changed from the key signal input circuit 7, and the signal is applied to the key signal correction circuit 5 through the control circuit 6, thereby making it possible to obtain images with different characteristics. Reference numeral 3 denotes an image boundary processing device, which has a function of smoothly expressing boundary portions using the contents of the key signal memory 2. For example, focusing on nine surrounding points including a certain center point over the entire screen, the value of the key signal is appropriately weighted to calculate the mixing ratio of the background image and the target image. Several thresholds are set for the ratio, the mixing ratio of each image is determined, and the density of the center point is calculated.

According to this method, adjacent points on the same image are given the same key signal as their surroundings, so they are not mixed with other images by mistake, and mixing is performed only on the boundary areas where the key signal changes. . As a result, processing is applied only to the boundary portion through the spatial filter. Therefore, attributes such as the shape of the image can be saved even for details of the image.

The image generating device 4 synthesizes the plurality of corresponding images in the image memory 1 by extraction or mixing according to the key signal output modulated by the image boundary processing device 3, and displays the composite on the image display device 10. . It consists of four image generators, an addition circuit, a logic circuit, a synchronization signal generation circuit, etc.
The R, G, and B values corresponding to each pixel are calculated and output according to the control signal.

The image memory 1. Key signal memory 22 Image boundary processing device 31 Image generation device 4. The key signal correction circuit 5, the control circuit 6, and the key signal input circuit 7 constitute an image composition device 8.

Effects of the Invention As is clear from the above description, the present invention provides an image memory, a key signal memory for storing a key signal which is a control signal for extracting a target image, and an image processing method using the contents of the key signal memory. an image boundary processing device that smooths images by blending each image at the boundary; and an image generation device that synthesizes and displays the contents of the image memory according to an output signal of the image boundary processing device; Since it is configured to combine images, it is possible to easily generate a high-quality composite image with less jaggies, which is an excellent effect. In addition, R, G of the image is used as control information for synthesis.
, Since the B value is not used directly and the key signal is created and processed, flexibility is increased, and the user can change the key signal from the key signal input circuit to the control circuit and the key signal correction circuit. Various visual effects such as changing the shape of the target image, changing the mixing ratio to blur it, or moving the position of the target image itself can be easily obtained. Further, the process of blending and blurring the corresponding images is applied only to the boundaries, which has the effect of preserving the detailed shapes without confusing the background with the target image. Such effects can be used not only in typical cases, such as compositing backgrounds and human figures, but also in text.

The same applies when displaying figures. By making the key signals multi-layered and setting them so that they can be identified, it is possible to simultaneously obtain various visual effects and a composite image of a large number of images.

[Brief explanation of drawings]

The drawing is a block diagram of main parts of an image synthesis device according to an embodiment of the present invention. 1...Image memo 1c 2...Key signal] Mori, 3...Image boundary processing device, 4...
...Image generator, 5...Key signal correction circuit,
6... Control circuit, 7... Key signal input circuit, q... Image synthesis device, 9... Image input device, 10...・Image display device.

Claims (3)

[Claims] (1) An image memory that stores multiple images, a key signal memory that stores a key signal that is a control signal for extracting a target image from among the image signals, and smooths the boundaries of a composite image based on the contents of the key signal memory. An image synthesizing device comprising: an image boundary processing device that generates a control signal to make a control signal; and an image generating device that synthesizes a plurality of images from the contents of the image memory and the output of the image boundary processing device. (2) The image boundary processing device has a function of extracting and smoothing only the boundary portion of the target image from the contents of the key signal memory, and acts as a spatial filter. ). (3) The image synthesizing apparatus according to claim (1), wherein the contents of the key signal memory are corrected to partially change the target image.