JPH036781A - Processing system for extraction of three-dimensional position information - Google Patents

Abstract

PURPOSE:To easily extract the 3-dimensional position information on an object by providing a coordinate transformation part which obtains the coordinates of residual three points when one of four feature points set on two frame images of two feature points is defined as an original point. CONSTITUTION:An image input part 101, a feature point extracting part 102, a coordinate transformation part 103, a reference point distance supply part 104, and a position information extracting part 105 are provided. When one of four characteristic points of two frame images of two feature points is defined as an original point, the coordinates of residual three points are used for extraction of a 3-dimensional position with the 3-dimensional position of the reference point of each feature point set as a standard. Thus it is possible to extract the 3-dimensional position information on the feature point of an object just by observing the moving object.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a three-dimensional position information extraction processing method performed by observing an object with one image input device. [Prior Art] One image input device In the conventional method of extracting three-dimensional position information about feature points on an object from a series of frame images obtained by observation with an image input device, a predetermined movement of the object is determined by a predetermined amount of movement. A stereo image is constructed from two input images obtained by observing the object before and after it is moved by the object moving means, and
Three-dimensional position information was extracted by dimensional position measurement.

[Problem to be solved by the invention]

Therefore, it is necessary to apply a predetermined movement to the object by a predetermined amount of movement, or to manually measure the amount of movement, which is a problem in that it takes time and effort to extract three-dimensional position information. was there.

In order to solve such conventional problems, the present invention has the following features:
The object of the present invention is to provide a three-dimensional position information extraction processing method that can easily extract three-dimensional position information of an object by observing a predetermined movement without setting any restrictions on the amount of movement of the object.

(Means for Solving the Problems) In the three-dimensional position information extraction processing method of the present invention, a series of frames obtained by observing a moving object with a single image input device such as a television camera are used. Out of a total of 4 feature points (2 feature points for each of the 2 frame images) from the image, calculate the coordinates of the remaining 3 points when one point is the origin, and calculate 1 of the 3 points. The feature is that the three-dimensional position of the feature point is extracted using one point as the reference for the three-dimensional position.

In the conventional case, the three-dimensional position information of each feature point was extracted from the coordinates of each feature point on two frame images using the principle of triangulation, so it was necessary to determine the amount of movement of the object in advance or to measure it separately. However, in the three-dimensional position information extraction processing method of the present invention, out of a total of four feature points on two frame images of two feature points, one point is set as the origin. It has a means for determining the coordinates of the remaining three points when It differs from conventional methods in the point of extraction.

(Embodiment) An embodiment of the present invention will be described below in detail based on one drawing. Fig. 1 shows a block diagram of a three-dimensional position information extraction processing method in this embodiment, and shows feature points. one of them
This is shown as an example in which the distances of two reference points to the image input device are given. In the figure, 101 is an image input section, 102 is a feature point extraction section, 103 is a coordinate conversion section, 104 is a reference point distance adding section, and 105 is a position information extraction section.

The image input unit 101 observes a moving object using an image input device such as a television camera, and outputs two frame images of a series of frame images obtained by observation to the feature point extraction unit 102. . The feature point extraction unit 102 extracts at least two feature points from the two frame images output from the 5-image human power unit 101, and extracts two or more feature points from the extracted feature points.
The coordinates on the two frame images are output to the coordinate conversion unit 103. Extraction of feature points from two frame images is as follows: 1. For example, points where the shape or surface color of an object change drastically (locally) are extracted by image processing, and points that are the same on the object are extracted. The associated points are defined as feature points. The coordinate conversion unit 103 converts two of the feature points output from the feature point extraction unit 102 into four feature points on the two frame images, the remaining three when one point is set as the origin.
The coordinates of the three points thus obtained are output to the digit 1 information extraction section 105. The reference point distance providing unit 104 provides a reference for the three-dimensional position of each feature point. For example, one point among the feature points is set as the reference point, and the distance from the reference point to the image input device is set as the S quasi. give. At this time.

If only the positional relationship between feature points is required, an appropriate value may be given as the distance of the reference point to the human-powered image device. The position information extraction unit 105 extracts the 3 output from the coordinate conversion unit 103.
The three-dimensional position information of the feature point in the two frame images is extracted and output from the coordinates of the two points and the distance of the reference point to the image human-powered device outputted from the reference point distance adding unit 104.

An example of the coordinate transformation unit 103 and the position information extraction unit 105 will be described below for a case where the movement of an object between two frame images does not include a rotational motion component.

If the camera model is represented by the central projection shown in Figure 2, the center of the camera lens is the origin, the optical axis of the camera is the Z axis, and the direction perpendicular to the image plane of the camera is the X axis. The three-dimensional position of Pk (Xk, Yk, Zk) is
Pk (7) 71z-L position on image (Uk, Vk)
, where r is the distance from the center of the camera lens to the image plane of the camera (r Zk=Zk (
It can be expressed as +).

As shown in FIG. 3, the points at the time of inputting the i-th and j-th frame images regarding the feature points PO and Pk are expressed as Pio, PjO and Pik, Pjk(k・
1..., N-1, N: number of feature points), the feature points on the two frame images output from the feature point extraction unit 102 are expressed as QiO(UiO, VIO), QjO(UjO, V
ja), Qik (Uik, Vik), Q
Let jk (Ujk, Vjk).

Point Q io on two frame images of two feature points.

Among the four points QjO, Qik, Qjk, 1, for example 2
When point Qjk is the origin, the remaining three points QiO, Qj
The coordinates after OQik transformation are (Wlx, Wly)
.

(W2x, W2y), (W3X, W3V), the converted coordinates of the three points output from the coordinate conversion unit 103 are expressed by equation (2).

QiO: (W3χ, W3y) − (UiO−Ujk, VIO−Vjk)QjO:
(W2x, W2y) - (UjO-Ujk, VjO-Vjk) Qik: (W
lx, Wly) −(Uik−Ujk, Vi
k-Vjk) (2) Next, the 1-position information extraction unit 105 will be explained. In FIG. 3, when the movement of the object between the two frame images does not include a rotational motion component, the following equation holds true. .

0Pjk = 0Pik + PjOPiO(
3) From equation (1) 1 equation (3).

The relationship is obtained, and the Z coordinate of the reference point that can be calculated by eliminating Zjk and f (by calculating from aZio using equation (6), i The three-dimensional positions of the feature points in the two frame images, the th and jth, can be determined manually.

From equations 0 (2), 1 (4), and 2 (5), the Z component of the three-dimensional position 1 of the feature point is:

W1xW2y-W2xW1y W1xW2y-W2xWIy Zjk= Zik+ ZjO-ZiQ
(6) can be expressed as the X component of the three-dimensional position of one feature point,
The Y component is.

It is obtained by substituting equation (6) into equation (1).

That is, the coordinates of the three points output from the coordinate conversion unit 103 are (Wlx, Wly), (W2X, W2
y).

(W3x, W3y) and the distance from the reference point PiO output from the reference point distance adding unit 104 to the image input device [
[Effects of the Invention] As explained above, according to the present invention, the coordinates of the remaining three points when one point is set as the origin among a total of four feature points on two frame images of two feature points. By using Three-dimensional position information of feature points on the object can be extracted simply by observing the moving object.

[Brief explanation of drawings]

Fig. 1 is a configuration diagram of a three-dimensional position information extraction processing method showing an embodiment of the present invention, Fig. 2 is a diagram showing a camera model, and Fig. 3 is a three-dimensional position change of feature points due to the movement of an object. FIG. 3 is a diagram showing the correspondence between two-dimensional positional changes of feature points on a frame image and a two-dimensional position change of a feature point on a frame image. 101... Image input unit, 102... Feature point extraction unit. 103... Locus 45 conversion section, 104...
Reference point distance adding unit 105...Position information extraction unit.

Claims (1)

[Claims] Observing a moving object using an image input device
In a three-dimensional position information extraction processing method that extracts dimensional position information, i in a series of frame images obtained from the image input device is
Feature points are extracted from each of the two frame images, the i-th and j-th frame images, and four It has a coordinate transformation means for determining the coordinates of the remaining three points when one point among the feature points is set as the origin, and using the coordinates of the three points determined by the coordinate transformation means,
A three-dimensional position information extraction processing method characterized by extracting a three-dimensional position of a feature point using one point among the feature points as a reference.