JPH0625648B2 - Stereoscopic method - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to a stereoscopic method for stereoscopically recognizing a three-dimensional object using a plurality of two-dimensional imaging means.

<Background of the Invention> In recent years, by observing an object from three directions with three TV cameras, points (hereinafter referred to as "characteristic points") that characterize the object such as a corner portion of the object are extracted, A method has been proposed in which three-dimensional coordinates are obtained and an object is three-dimensionally recognized (Nikkei Mechanical, January 2, 1984 issue). This method uses the epipolar line on the screen of the TV camera.
e) is obtained, the object point image located on this line is detected, the object point images on the respective screens are associated with each other, and the three-dimensional coordinates of the subsequent object point are calculated. However, in the case of this type of method, due to errors etc. that occur at each stage of image processing,
The object point image that should be originally detected as the corresponding point may be displaced from the epipolar line, and in such a case, the object point image is removed from the corresponding point candidates, which causes a decrease in object recognition accuracy. There was a problem.

<Object of the Invention> The present invention provides a novel method for extracting corresponding object point images without omission, and an object thereof is to improve the stereoscopic recognition accuracy of an object.

<Structure and Effect of the Invention> In order to achieve the above object, in the present invention, a band-shaped region having a constant width is set in an epipolar line, an object point image included in this band-shaped region is extracted, and an object point on each image is extracted. We decided to associate the images.

According to the present invention, even if the object point image to be detected as the corresponding point is located from the epipolar line, the object point image can be reliably extracted, contributing to the improvement of stereoscopic recognition accuracy, etc.
The remarkable effect of achieving the object of the invention is achieved.

<Description of Embodiments> FIG. 1 shows three television cameras 1, 2, 3 on a fixed machine base 4.
Shown below is a stereoscopic device 5 in which (first camera 1, second camera 2, and third camera 3) are arranged. An image of an object obtained by each television camera is taken into the image processing device 6,
A series of stereoscopic recognition processes such as association of object images and calculation of three-dimensional coordinates of object points are executed.

FIG. 2 is a principle diagram showing a method of associating the object point images between the images.
0, 30 (hereinafter, referred to as the first image 10, the second image 20, and the third image 30) on the object point image P ₁ of the feature point P,
P ₂ and P ₃ are shown. Also On the second image 20, the image of the straight line F ₁ P ₁ connecting the focus F ₁ and the object point images P ₁ of the first camera 1 (the linear images of the epipolar line) l ₂ is set, similarly On the third image 30, epipolar lines l ₃ and m ₃ of the straight line F ₁ P ₁ and the straight line F ₂ P ₂ are set.

FIGS. 3 (1) (2) (3) show the images 10, 20, 30 described above. According to the figure, the object point image P ₂ in the second image 20 is located on the epipolar line l ₂ , and the object point image P ₃ in the third image 30 is located on the intersection of the epipolar lines l ₃ and m _3. . Therefore, the object point images P ₁ , P ₂ and P ₃ are the feature points P.
It is understood that they are points that correspond to each other as images of
Therefore, the three-dimensional coordinates of the feature point P are straight lines F ₁ P ₁ , F ₂ P ₂ and F.
It can be obtained as the intersection of ₃ P ₃ . Figure 3 (2) (3)
2 also shows object point images R ₂ and R ₃ of other feature points R located on the extension of the straight line F ₁ P in FIG. ₂ , in which case the object point image R ₃ is the epipolar line l ₃ Not located on the intersection of m ₃ .

FIG. 4 shows the features of the stereoscopic viewing method according to the present invention. The epipolar line l ₂ of the second image 20 and
With respect to the epipolar lines l ₃ and m ₃ of the third image 30, band-shaped regions L ₂ , L ₃ and M ₃ each having a constant width and including each line in the center are set. The present invention, in the associating process object point image, when the object point images are contained within these band region _{L 2, L 3, M 3} , the object point image epipolar line l _2, l _3, m ₃ above It is characterized in that the process is considered to be located at.

Now, focusing on the object point image P ₁ on the first image 10, first, the epipolar lines l ₂ and l ₃ generated by the object point image P ₁ on the second image 20 and the third image 30 are obtained, and
Band regions L ₂ and L ₃ having a predetermined width are set for each epipolar line l ₂ and l ₃ . Next, in each of the images 20 and 30, a set of object point images included in the strip regions L ₂ and L ₃ is extracted by the method described below.

Now, when the ij Cartesian coordinate system is set for each image 20, 30, the equation of the epipolar line is ai + bj + c = 0,
Further, if the coordinates of the object point image are (i ₀ , j ₀ ), the distance d from the object point image to the epipolar line can be obtained by the following equation.

And when the width 1/2 of the band-shaped area is set to the threshold value TH,
When the position of the object point image satisfies the following equation, it can be determined that the object point image is included in the band-shaped region.

d <TH ..... Thus, the calculation of the formula is executed for all the object point images on the first and second images 20 and 30, and the formula is determined, so that each band-shaped region L ₂ and L ₃ is detected. Find the set of object point images included.
Next, for each object point image included in the strip-shaped region L ₂ of the epipolar line l ₂ , an epipolar line (epipolar line m ₃ is one of them) generated on the third image 30 is obtained, and the epipolar line is similarly obtained. (for epipolar line m ₃ band-like region M ₃₎ band-shaped region for setting the. Then, because of the object point image on the third image 30,
The calculation of the formula and the determination of the formula are executed to obtain a set of object point images included in each band-shaped region. Then, the set of object point images in the strip-shaped region L ₃ is collated with the set of object point images in each strip-shaped region, and the object point images commonly included in both regions (in this case, L ₃ and M ₃ ). P ₃ is obtained and the object point image P of the third image 30 is obtained.
_3, extracts an object point image P ₂ of the second image 20, and the object point images P ₁ of the first image 10 as a corresponding point.

[Brief description of drawings]

FIG. 1 is a front view showing a configuration example of a stereoscopic device, FIG. 2 is an explanatory view showing the principle of a method of associating object point images, and FIGS. 3 and 4 are explanatory views showing images of each TV camera. Is.

Claims (1)

[Claims] 1. An image of an object is obtained by at least three two-dimensional image pickup means, epipolar lines are provided on these images, and a band-shaped region having a constant width is set in each epipolar line, and then the epipolar line is included in the band-shaped region. A stereoscopic method comprising extracting an object point image and associating the object point image on each image with each other.