KR100189488B1 - Method for converting digital stereoscopic image - Google Patents

Abstract

The present invention relates to a digital stereoscopic image conversion method, and in particular, separating a left image and a right image from a 3-D image obtained by synthesizing a left and right image by a line interlacing method; Restoring each of the separated left and right images to a complete image by a line compensation method; Converting the reconstructed left and right images to the RGB method; Extracting only a red image from one of the converted RGB left and right images to generate a red image, and simultaneously removing the red image from the other image to generate a blue image; And outputting a stereoscopic RGB image by synthesizing the generated red image and blue image.

Therefore, in the present invention, by converting the stereoscopic image obtained by the binocular stereoscopic camera to the stereoscopic image of the red-blue method, more users can enjoy the stereoscopic image cheaply and easily.

Description

How to convert digital stereoscopic video

1 is a schematic diagram illustrating a general red-blue stereoscopic image.

2 is a schematic diagram illustrating a general liquid crystal shutter stereoscopic image.

3 is a schematic diagram illustrating a digital stereoscopic image conversion method according to the present invention.

4 is a diagram illustrating an embodiment of separating left and right images from a liquid crystal shutter type stereoscopic image according to the present invention.

The present invention relates to a digital stereoscopic image conversion method, and more particularly, to a method of converting a stereoscopic video picked up by a binocular stereoscopic camera to a digital stereoscopic video of a red-blue method.

The three-dimensional sense is perceived when both eyes of the person look at the real world because of the different images on the retina due to the distance between the two eyes (2.5 inches). The inconsistency of the image that occurs in the retina refers to the horizontal distance to the same target point in the image projected onto the retina.

Parallax, on the other hand, is an important factor in determining the width of the discrepancy in referring to the horizontal distance between two points of the image projected on the monitor.

Positive parallax is when the observation point is farther than the stereoscopic plane and is called positive parallax. It occurs when the parallax is less than or equal to the distance between the eyes.

Zero parallax refers to the case where the observation point is at the same distance as the solid plane. In other words, having a parallax of zero means that there is no space between the separated images, and thus the three-dimensional feeling is not felt. Negative parallax refers to a case closer to the three-dimensional plane and occurs when the line of sight intersects, giving a sense of protruding three-dimensionality.

Research into realizing stereoscopic images using the characteristics of the human eye as described above has been steadily studied for a long time. Among many studies, binocular stereoscopic images using glasses are generally commercialized and widely used.

The most common technologies include red-blue glasses and liquid crystal shutter glasses. Referring to FIG. 1, the red-blue method is a video signal synthesized by synthesizing a red image (R) filtered from a left image and a blue image (B + G) filtered from a right image into a positive or negative parallax image. This allows the viewer to enjoy a stereoscopic image by filtering through a red filter on the left eye and a blue filter on the right eye so that only the red image is visible to the left eye and only the blue image to the right eye. Until now, the red-blue stereoscopic image has been read by a scanner that uses a film-type stereoscopic camera to make the above conversion process, or by computer animation or by recording two-dimensional images with two film cameras. Through the analog way to realize stereoscopic through it was able to enjoy stereoscopic video. Such a red and blue method has a disadvantage in that color reproduction is poor while a user can enjoy a stereoscopic image cheaply.

Referring to FIG. 2, in the liquid crystal shutter method, the left image is an odd line, the right image is an even line, and a single stereoscopic image is synthesized and displayed on the monitor, and the viewer is synchronized with the image displayed on the monitor. In this case, the right shutter is closed and the left shutter is opened to transmit the image only to the left eye, and when the right image is closed, the left shutter is closed and the right shutter is opened to transmit the image only to the right eye so that the stereoscopic image can be enjoyed. Therefore, this method can provide a perfect stereoscopic effect and overall clear picture quality, but there is a problem in that a circuit configuration for switching the liquid crystal shutter at high speed in synchronization with the image and the liquid crystal shutter glasses 20 are expensive. In addition, a dedicated interface board for connecting the computer and the glasses must be provided. Accordingly, for a stereoscopic image environment using a computer, the stereoscopic image may not be enjoyed unless a specially manufactured liquid crystal shutter glasses and a dedicated interface board for synchronization are provided.

In recent years, as various users of the computer system and the international PC communication network such as the Internet has been commercialized and its users are rapidly expanding, various various multimedia services are performed. Among them, the spread of 3-D video is gradually increasing, but as described above, there is a problem in that it cannot be enjoyed unless equipment for each format is provided because of different stereoscopic video formats.

Accordingly, an object of the present invention is to provide a method for converting a three-dimensional stereoscopic video to a red-blue stereoscopic video so that more users can easily enjoy the three-dimensional image in order to solve the problems of the prior art.

In order to achieve the above object, the method of the present invention comprises the steps of: separating the left image and the right image from the 3-D digital image synthesized from the left and right images by line interlacing; Restoring each of the separated left and right images to a complete image by a line compensation method; Converting the left and right images to an RGB method; Extracting only a red image from one of the converted RGB left and right images to generate a red image, and simultaneously removing the red image from the other image to generate a blue image; And outputting a stereoscopic RGB image by synthesizing the generated red image and blue image.

In addition, the modified method of the present invention comprises the steps of acquiring left and right images from two cameras having parallax for stereoscopic images, respectively; Converting the obtained left and right images into an RGB image; Extracting only a red image from one of the converted RGB left and right images to generate a red image, and simultaneously removing the red image from the other image to generate a blue image; And outputting a stereoscopic RGB image by synthesizing the generated red image and blue image.

Hereinafter, with reference to the accompanying drawings will be described in more detail the present invention.

3 schematically shows a stereoscopic image conversion method according to the present invention. The left image 32 and the right image 34 are separated from the 3-D image 30 in which the left and right images are synthesized by the line interlacing method, and each of the separated left image 32 and the right image 34 is line compensated. The method restores the complete image. The restored left and right images are converted into an RGB method, and only a red image is extracted from the converted left image to generate a red image 36. At the same time, a red image is removed from the RGB right image to generate a blue image 38. The red image 36 and the blue image are synthesized to output a stereoscopic R'GB image 40.

The line compensation method compensates a line by using simple copying of pixel information on an upper or lower line, or average value information or statistical characteristics of pixels corresponding to the upper and lower pixels and the actual image existing on the left and right sides of the upper and lower pixels. .

4 shows an embodiment of the line compensation method according to the present invention. That is, the left image composed of the odd lines of the line interlaced stereoscopic image is separated, and the even line of the left image obtains an average value of two pixel informations of the upper and lower lines to restore the reconstructed region. Similarly, the right image composed of even lines is separated, and the odd line of the right image is obtained by reconstructing the reconstructed region by calculating an average value of two pixel information of upper and lower lines.

Before dividing the left and right images, it is preferable to adjust brightness, contrast, color, etc. for the purpose of adjusting vertical and horizontal alignments or converting colors for three-dimensional conversion of left and right images for 3-D.

In addition, the screen may be enlarged or reduced at any one of the above steps.

In contrast to the above-described embodiment, when the red image is removed from the left image and the red image is extracted from the right image, the color filter of the red and blue glasses should also be left and right turned blue.

The conversion method described above is for the case of the line interlaced stereoscopic images taken with a 3-D camera, and when the stereoscopic images are taken with two general cameras, each of them is picked up from two cameras having a parallax for stereoscopic images. Input left and right images, convert the input left and right images into RGB images, extract only red images from the converted RGB left images, generate red images, and simultaneously remove red images from RGB right images By generating an image, synthesizing the generated red image, and outputting a stereoscopic RGB image, a red-blue stereoscopic image can be easily made.

As described above, in the present invention, a user who enjoys a stereoscopic image using a computer can easily convert the stereoscopic image of different formats to a red-blue method and enjoy it even if the user has only a simple red-blue filter without special glasses for stereoscopic images. This stereoscopic image can be used easily.

Claims (5)

Separating a left image and a right image from the 3-D image obtained by synthesizing the left and right images by a line interlace method; Restoring each of the separated left and right images to a complete image by a line compensation method; Converting the reconstructed left and right images to the RGB method; Extracting only a red image from one of the converted RGB left and right images to generate a red image, and simultaneously removing the red image from the other image to generate a blue image; And outputting a stereoscopic RGB image by synthesizing the generated red image and blue image. The method of claim 1, wherein the line compensation method uses a simple copy of pixel information on an upper or lower line, or average value information or statistical characteristics of pixels corresponding to the upper and lower pixels and the actual image existing on the left and right sides of the upper and lower pixels. Digital stereoscopic image conversion method, characterized in that for compensating for the line. According to claim 1, Brightness, contrast, color, etc. are adjusted for the purpose of adjusting the vertical and horizontal alignment or converting the color for the stereoscopic conversion of the left and right images for 3-D before separating into the left and right images Digital stereoscopic image conversion method characterized in that. The method of claim 1, wherein the screen is enlarged or reduced at any one of the steps. Inputting left and right images respectively picked up from two cameras having parallax for stereoscopic images; Converting the input left and right images into an RGB image; Extracting only a red image from one of the converted RGB left and right images to generate a red image, and simultaneously removing the red image from the other image to generate a blue image; And outputting a stereoscopic RGB image by synthesizing the generated red image and blue image.