WO2009002327A1

WO2009002327A1 - Optical rangefinder for a 3-d imaging system

Info

Publication number: WO2009002327A1
Application number: PCT/US2007/015302
Authority: WO
Inventors: Youngshik Yoon
Original assignee: Thomson Licensing
Priority date: 2007-06-28
Filing date: 2007-06-28
Publication date: 2008-12-31

Abstract

A 3-D imaging system including an optical rangefinder attached to a camera is described. The optical rangefinder is used to convert un-superimposed images into fine object depth information based on a comparison with a reference point. Additionally, full 3-D image capturing is obtained using different infrared wavelengths for object differentiation.

Description

OPTICAL RANGEFINDER FOR A 3-D IMAGING SYSTEM

FIELD OF THE INVENTION

The invention relates generally to 3-D image capturing. More specifically, the invention relates to an optical rangefinder for a 3-D imaging system.

BACKGROUND OF THE INVENTION

It is well known that an image may be enhanced with an appearance of depth by converting the image into a so-called 3-D image. This is most often accomplished by optically polarizing the images which are to be viewed by a viewer's left eye differently than the images which are to be viewed by a viewer's right eye. The 3-D effect is perceived by the viewer when the viewer views the polarized images though the use of polarized filter lenses, commonly configured as '3-D viewing glasses' with a polarized filter for use with the left eye of the viewer and a differently polarized filter for use with the right eye of the viewer.

Typically, depth information for 3-D image capturing is obtained using the human eye and a laser along with a detecting device including an infrared sensor and optics or using a stereo camera. However, the use of the laser has often been objected to because of safety reasons, and also using a stereo camera requires high cost and causes often limited view. In particular, the possibility that the laser will shine in the human eyes directly causing irreparable damage. Thus, a method of obtaining depth information for 3-D image capturing at low cost without using the human eye is needed.

SUMMARY OF THE INVENTION

A 3-D imaging system includes an optical rangefinder attached to a camera.

The optical rangefinder is used to convert un-superimposed images into fine object depth information based on a comparison with a reference point. Additionally, full 3- image capturing is obtained using different infrared wavelengths for object fferentiation.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described with reference to the accompanying figures of which.

FIG. 1 is a schematic drawing of an optical rangefinder of the present invention; FIG. 2 shows the optical rangefinder of FIG. 1 attached to a camera of a 3-D imaging system; and

FIG. 3 is a flow chart describing the conversion of image information into depth information.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is directed to a 3-D imaging system including an optical rangefinder attached to a camera. The optical rangefinder is used to convert un- superimposed images into depth information based on a comparison with a reference point. Additionally, full 3-D image capturing is obtained using different infrared wavelengths for object differentiation.

Referring now to FIG. 1 , there is shown a schematic drawing of an optical rangefinder 10 of the present invention. The optical rangefinder includes a roof pentaprism 25, a beam splitter 12, an eye piece lens 14, mirrors 17, 18, and a converging lens 16. Un-superimposed image information of an object is input to the optical rangefinder 10 at two points 21, 20.

Image information 21 from a first perspective of the object is provided directly to the human eye 11 through the beam splitter 12. Image information 20 from a second perspective of the object is provide to the human eye 11 via the roof pentaprism 25 converging lens 16 and eyepiece lens 14. The roof pentaprism 25 inverts the image 20. The converging lens 16 focuses the inverted image and shifts it to superimpose the images 20, 21 as the lens 16 is adjusted. In order to uperimpose the images 20, 21 the converging lens 16 should rotate. Then, based n the pivoting angle of the lens 16 a distance relationship can be established.

In operation the optical rangefinder 10 is attached to a camera 5, as shown in FIG. 2a. Referring to FIG. 3, initial characteristics of the optical rangefinder 10 are established as indicated in step 100. The initial characteristics of the optical • rangefinder 10 include determining the distance D (FIG. 2a) between the two apertures 20, 21 (FIG. 1).

Thereafter, the distance from the viewing point 20, 21 to the reference point 35 should be determined, as indicated in step 200. The reference point 35 is the point at which perfect superposition occurs. This distance may be determined, for example, by attaching an infrared radiator to an object at the reference point and detecting the distance the infrared light travels or by simple triangular geometry calculation. For multiple objects, several infrared radiators, each with a different frequency, may be used to differentiate between them. Referring to step 300, random objects A 40, B 50, C 60 are placed between the optical rangefinder 10 and the reference point 35. Then, based on the distance between the two apertures 20, 21 as well as the distance from the viewing point to the reference point 35 the horizontal distance between the two un-superimposed images can be calculated to find the vertical position A 40, B 50, C 60 of the random objects, using simple geometry i.e., using parallax. This establishes the distance of each object A 40, B 50, C 60 from the viewing point 35.

Referring to FIG. 2b, as the un-superimposed images of the random objects A 40ai, A 4Oa₂, B 5Ob₁, B 50b2, C 60ci, C 6OC₂ get closer as the distance between them gets closer to the reference point 35. Thus, the horizontal distance 59 between un-superimposed images C 60c-i, C 6Oc₂ is greater than the horizontal distance 49 between un-superimposed images B 5Ob₁, B 5Ob₂. Similarly, both horizontal distances 59, 49 are greater than the horizontal distance 39 between un- superimposed images 4Oa₁, A 4Oa₂.

Once the distance of each object A 40, B 50, C 60 to the viewing point 35 is calculated the acquisition of depth information across the field may be similarly calculated as indicated in step 400 of FIG. 3. A great advantage of such depth information is to be continuous, not discreet, i another word, defining depth levels is up to user. Therefore, it can be applied to any 3D display with different depth levels.

3-D image generation of auxiliary objects, such as for example, a hair, which may be attached to the main object can also be calculated as indicated in step 310. This is because the 2-D image will be recorded by the camera and once the initial depth information is calculated, follow the calculation from the main image.

For off-axis objects the same method of determining the depth information may be used. Just the different angular values across the field must be considered when calculating the depth of an object.

The foregoing illustrates only some of the possibilities for practicing the invention. Many other embodiments are possible within the scope and spirit of the invention. It is, therefore, intended that the foregoing description be regarded as illustrative rather than limiting, and that the scope of the invention is given by the appended claims together with their full range of equivalents.

Claims

1. A 3D image system, comprising: a camera; and a detecting device coupled to the camera, wherein distance to an object is determined based on a comparison of un-superimposed images with a reference point.

2. The 3D image system of claim 1 wherein the detecting device includes an infrared detector.

3. The 3D image system of claim 1 wherein the detecting device includes a roof pentaprism.

•4. The 3D image system of claim 1 wherein the detecting device includes a converging lens.

5. A method of converting image information to depth information in a 3-D imaging system, comprising the steps of: determining a distance from a viewing point to a reference point; and converting a horizontal distance between un-superimposed images into depth information based on a comparison with the distance to the reference point.