CN1176351C - Method and device for dynamic multi-resolution three-dimensional digital imaging - Google Patents

Abstract

The invention discloses a dynamic multi-resolution three-dimensional digital imaging method and device, which include a double acousto-optic lighting emitter, an image sensing receiver and an image processor. Structured light sequences with different spatial periods or frequencies and different phase shifts are generated by the said illumination emitter, and are projected onto the digitized scene in time series, and the depth information of the scene is encoded in a space-time sequence, which is received by the image sensor The digital image processor obtains the digital image of the structured light that has undergone sequence space-time coding, and the image processor obtains the multi-resolution 3D depth information of the scene by decoding the space-time sequence coded structured light image, that is, the 3D digital image. The advantages of the present invention are: the device has no mechanical movement, good stability and repeatability; the double acousto-optic modulation lighting emitter adopts a symmetrical structure of "common optical path" or "equal optical path", and has strong anti-interference performance; spatial structured light real-time Frequency conversion and phase modulation, hierarchical space-time phase reconstruction, high precision and flexibility of 3D digital imaging.

Description

Method and device for dynamic multi-resolution three-dimensional digital imaging

Technical field

The invention relates to a dynamic multi-resolution three-dimensional digital imaging method and device, belonging to the three-dimensional digital imaging technology.

Background technique

The 3D imaging technology based on active laser triangulation can obtain the high-density depth data of the scene without contact, including the 3D imaging technology of laser point scanning triangulation and laser sheet scanning triangulation, such as CYBERWARE ( www.cyberware.com ) and DIGIBOT ( www.digibotics.com ) company's three-dimensional digitizer, said method and device all need mechanical or optical scanning mechanism, so the speed of three-dimensional imaging is greatly limited. The 3D imaging technology based on fringe projection and phase measurement does not require any mechanical or optical scanning mechanism. It not only has the characteristics of non-contact, high-density depth data but also has the characteristics of full-field depth information acquisition, so it has been widely researched and developed.

The currently proposed 3D digital imaging based on fringe projection and phase measurement generally includes two categories: laser interference fringe projection and white light grating projection. Stripes or gratings are projected onto the scene as a single-frequency space carrier, and the scene encodes depth information in the modulated space carrier. The carrier image is optically or digitally decoded to obtain the depth data of the scene. One of the efficient ways to decode the modulated space-carrier image is through the "phase-shifting algorithm", which requires the generation of more than three phase-shifted encoded space-carrier images. Most of the methods mentioned can only produce fixed and single spatial frequency (or period) fringes or gratings, it is difficult to generate phase shift quickly (such as video rate), and cannot be applied to the occasion of dynamic three-dimensional sensing. Three-dimensional imaging is not universal. For example, when the scene has a large gradient change, all the methods mentioned above will fail.

Comparable technical literature has the following four:

[1] Invention patent: publication number CN1205453A

[2] Invention patent: authorized announcement number CN1093935C.

[3] M.Rioux, "Three dimensional imaging method and device". US Patent 4,627,734

[4]M.Halioua and V.Srinivasan, "Method and apparatus for surface profilometry".US Patent4,641,972

Contents of Invention

The purpose of the present invention is to provide a dynamic multi-resolution three-dimensional digital imaging method and device. This technology uses sinusoidal space carrier sequence lighting with space conversion and phase shift in real time, and is applied to dynamic three-dimensional sensing occasions. It has universal applicability to three-dimensional imaging of objects of any shape.

In order to achieve the above object, the present invention is achieved through the following technical solutions. A device consisting of a dual sound and light lighting transmitter, an image sensor receiver and an image processor is used to implement a dynamic multi-resolution three-dimensional digital imaging method for the scene. In this method, the double acousto-optic modulation lighting emitter and the image sensing receiver are placed on the same plane in space to form a fixed angle, and they form a triangle with the illuminated scene, and the line connecting the relative positions is the baseline, which is characterized in that: The transmitter illuminates the scene with the sinusoidal space carrier sequence of space frequency and phase shift generated in real time, and encodes the space and time sequence of the depth of the scene, that is, ξ ₁ , ξ ₂ = xξ ₁ , ... ξ _M = xξ _M-1 Different frequencies and the same frequency use δ ₁ , δ ₂ ... δ _N different phase-shifted fringe structured lights to illuminate the scene sequentially, and M×N coded fringe intensity distribution maps are obtained respectively, and the "phase-shift algorithm" is used to calculate each The phase diagrams Φ ₁ , Φ ₂ , Φ ₃ ... Φ _M of the fringe intensity distribution diagram, eliminate 2π uncertainty in each phase diagram, so as to obtain depth images with progressive resolutions for the scene, and obtain space and time series The encoded phase-shifted sinusoidal carrier structured light is then decoded to obtain multi-resolution 3D depth information of the scene, that is, a 3D digital image.

According to the above-mentioned method, the device for realizing dynamic multi-resolution three-dimensional digital imaging includes a dual acousto-optic modulation illumination transmitter, an image sensing receiver and an image processor, and said image sensing receiver includes an optical imaging lens and a photoelectric The optical imaging lens can be an imaging lens or lens group with a fixed focal length or a variable focal length, a binary optical imaging system, a diffraction element imaging system, and a microscopic imaging system; the photodetection device can be a charge-coupled device, a liquid crystal device, a spatial light modulation device or a digital camera; the image processor is a combination of a digital signal processor and a programmable application-specific integrated circuit, or a combination of a general-purpose image processing card and a computer. It is characterized in that the dual acousto-optic Modulated lighting transmitter, which is composed of coherent light source, beam splitter, dual acousto-optic modulator and radio frequency drive circuit, direction trimmer, spatial filter, beam spacing converter (optical wedge), focusing lens, and microscope objective lens ; Its driving radio frequency signal is a driving signal with the same frequency and a controllable phase difference of 0-2π, and the frequency changes in real time.

The advantages of the present invention are: (1) the overall system adopts an all-solid-state structure without any mechanical movement, so the stability and repeatability are very good; "equal optical path" symmetric structure, so it can compensate the influence of air disturbance; (3) Through different combinations of two or more double acousto-optic modulation illumination emitters and image sensing receivers in different spatial positions, fringe structures in different orientations can be obtained Light sequences, thereby greatly increasing the precision and flexibility of 3D digital imaging.

                                Explanation of the attached drawings:

Fig. 1 is the working principle schematic diagram of the inventive method

Figure 2 is a block diagram of a dual acousto-optic (AOM) modulated lighting transmitter

Fig. 3 is a schematic diagram of the position arrangement when a dual acousto-optic modulation illumination transmitter and two image sensing receivers are used to illuminate and image a scene.

Fig. 4 is a top view of the device in Fig. 3 when placed in a straight line on the same plane.

Fig. 5 is a top view of the device in Fig. 3 when placed in a triangle on the same plane.

Fig. 6 is a schematic diagram of the placement of an image sensing receiver and two dual acousto-optic modulation illumination transmitters for imaging the scene.

Fig. 7 is a top view of the device in Fig. 6 when placed in a straight line on the same plane.

Fig. 8 is a top view of the device in Fig. 6 when placed in a triangle on the same plane.

In the figure, 101 is a dual acousto-optic modulation lighting transmitter, 102 is an image sensing receiver, 103 is an image processor, 104 is a scene, 201 is a coherent light source, 202 is a beam splitter, 203 and 204 are acousto-optic modulators, 205 and 206 are direction trimmers (optical wedges), 207 is a dual spatial filter, 208 is a beam combiner, 209 is a focusing lens, and 210 is a microscope objective lens.

Detailed ways

Embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

As shown in Figure 1: the dual acousto-optic modulation illuminating emitter and the image sensing receiver form a fixed angle on the same plane in space, the optical axis of the illuminator and the imaging optical axis and the distance between the illuminating emitter and the image sensing receiver The relative positions form a triangle. The frequency and phase of the radio frequency driving signal of the double acousto-optic modulation illuminating emitter are controlled by the computer of the image processor or the digital signal processor and the D/A conversion. The connection line between the relative positions of the lighting transmitter and the image sensing receiver is defined as the baseline, and the double acousto-optic modulation lighting transmitter and the image sensing receiver are placed in front of the scene to be measured together. The dual acousto-optic modulation lighting emitters generate time-space stripe structured light sequences with spatial frequency conversion and different phase shifts, and project them onto the scene. Spatial phase encoding and transfer of the result to the image processor. The specific operation adopts the following steps: (1) Use the fringe structure with the lowest spatial frequency ξ ₁ (for example, ξ ₁ is the reciprocal of a spatial fringe period) and different phase displacements (δ ₁ , δ ₂ ,..., δ _N , N≥3) The light sequentially illuminates the measured scene, and the imaging system and the image sensing receiver sequentially acquire the fringe intensity distribution map of each phase-shifted space code, and use the "phase-shift algorithm" to calculate the phase map φ ₁ of the fringe intensity distribution (without 2π uncertainty), according to the above-mentioned triangular relationship, the non-integer fringe order can be determined by the calculation of the phase diagram, thereby further obtaining the rough outline of the scene; (2) adjust the spatial frequency to ξ ₂ =Xξ ₁ , where X is a positive integer, the measured scene is illuminated by fringe structured light with frequency ξ ₂ and different phase shifts, the second phase map φ ₂ is calculated according to the method mentioned above, and the 2π uncertainty can be eliminated by using the phase recovery algorithm, and we get The phase diagram of _φ2 will increase the resolution of φ2 by X times compared with that of φ1, because the 2π uncertain result that has been resolved by _ξ1 in the previous time will be inherited in the process of measurement by _ξ2 ; (3) once If ξ ₂ distinguishes 2π uncertainty, then the third resolution level measurement can be carried out. Repeating this can continuously improve the resolution of the three-dimensional imaging of the outline of the scene.

FIG. 2 is a specific embodiment of a dual acousto-optic (AOM) modulated illumination emitter 101 . The light beam generated by the coherent light source 201 passes through the dichroic prism 202 to form two beams of parallel light, which are respectively incident on the input ends of the two acousto-optic modulators 203 and 204 at the Bragg angle. Take out the positive first-order diffracted light, readjust it into two beams of parallel light through two optical wedges 205 and 206, and then enter the beam combiner 208 through a spatial filter to change the distance between them, and then pass through the focusing lens 209 to obtain the microstructure of the interference field, and then The enlarged sinusoidal fringe distribution is obtained through the microscopic amplification unit 210 . The frequency of the RF driving signals of two AOMs can be controlled by a computer to change the diffraction angle of the first-order diffracted light of the two AOMs sequentially, and the phase difference of the RF driving signals of the two AOMs can be controlled. The phase shifts between the first-order diffracted lights of two AOMs can be changed in time sequence, thereby generating space-time fringe structured light sequences with spatial frequency conversion and different phase shifts. Both the fringe spatial frequency and the phase shift amount can be changed by changing the bandwidth of the radio frequency signal, and the delay of changing the sound wave propagation can be realized on the order of milliseconds. Therefore, the time-space frequency conversion fringe pattern illumination generated by the above method can be much higher than the video rate ( about 40ms per frame).

Fig. 3 is another kind of embodiment of the present invention, and it is to adopt a dual acousto-optic modulation photographic transmitter and two image sensing receivers to illuminate the scene and place the schematic diagram of imaging device arrangement, Fig. 4 and Fig. 5 are Fig. 3 The top view of the two spaces with different arrangements. In this embodiment a dual acousto-optic (AOM) modulated illumination emitter 101 and two image sensing receivers 102 are used, from which the spatially variable frequency and out-of-phase The time-space fringe structured light sequence of the shift amount, the fringe spatial frequency and the phase shift amount can be changed by changing the bandwidth of the radio frequency signal, and the two image sensing receivers acquire the passing scene sequentially or simultaneously from both sides The light intensity distribution of the time-space sequence fringe pattern encoded by the depth information, and the result is sent to the image processor for phase decoding of the light intensity distribution of the encoded time-space sequence fringe pattern, and the computer will correspond to the two The two sets of phase decoded data from two receivers are fused together to obtain the final result. The use of two image sensing receivers can avoid the occlusion phenomenon caused by the depth of the scene, and increase the field of view range of the three-dimensional sensing.

Fig. 6 is still another embodiment of the present invention, and it is to adopt an image sensing receiver and two double acousto-optic irradiation emitters to carry out the schematic diagram of the location arrangement of the imaging device to irradiate the scene, and Fig. 7 and 8 are them in two kinds of spaces Top view of different arrangements. In this embodiment two dual acousto-optic (AOM) modulated illumination emitters 101 and an image sensing receiver 102 are used to generate and project the Two space-time fringe structured light sequences with the same or different spatial orientation and spatial frequency and different spatial frequency and phase shift, the spatial frequency and phase shift of the fringe can be changed by changing the bandwidth of the radio frequency signal. The sensing receiver obtains the light intensity distribution of the time-space sequence fringe pattern corresponding to the two double acousto-optic (AOM) modulated lighting emitters after being encoded by the depth information of the scene sequentially or simultaneously, and transmits the result Until the image processor performs phase decoding on the light intensity distribution of the encoded time-space sequence fringe pattern sequentially or simultaneously, the computer fuses the two sets of phase decoding data corresponding to the two emitters to obtain the final result. Using two dual acousto-optic (AOM) modulated illumination emitters can generate sequential space-time structured light illumination with the same or different spatial orientation and frequency, which can greatly increase the sensitivity and universality of 3D digital imaging.

Claims (2)

1, a kind of method of 3 D digital imaging of dynamic multiple resolution ratio, this method is placed on fixed angle of formation in the same plane, space with alliteration optical modulation illumination transmitter and image sensing receiver, they and illuminated scenery constitute a triangle, the line of relative position is a baseline, it is characterized in that: the illumination transmitter is with the spatial variable frequency of real-time generation and the sine space carrier wave sequence illumination scenery of phase shifts, and scenery carried out the room and time sequential coding of the degree of depth, promptly be respectively with ξ ₁, ξ ₂=x ξ ₁... ξ _M=x ξ _M-1Different frequency and same frequency are with δ ₁, δ ₂δ _NOut of phase is moved the striated structure light scenery that throws light on successively, obtains the fringe intensity distribution plan of M * N coding respectively, utilizes " phase shift algorithm " to calculate the phase diagram Φ of each fringe intensity distribution plan ₁, Φ ₂, Φ ₃Φ _MIn each phase diagram, eliminate 2 π uncertainties, thereby scenery is obtained the degree of depth picture that resolution is gone forward one by one successively, obtain the structured light of the phase shift sinusoidal carrier of room and time sequential coding, obtain multiresolution three-dimensional depth information to scenery, i.e. 3-dimensional digital picture through decoding again.

2, in accordance with the method for claim 1 a kind of, realize the device of the 3 D digital imaging of dynamic multiple resolution ratio, this device comprises that alliteration optical modulation illumination transmitter, image sensing receiver and image processor constitute, said image sensing receiver comprises that optical imaging lens and photodetector constitute, optical imaging lens can be imaging len or the lens combination that focuses distance or varifocal, the binary optical imaging system, diffraction element imaging system, micro imaging system; Said photoelectric detector can be a charge-coupled image sensor, liquid crystal device, spatial light modulation device or digital camera; Described image processor is the combination of digital signal processor and programmable asic, also can be that general image transaction card and computer combined constitute, it is characterized in that: alliteration optical modulation illumination transmitter, it is to be made of coherent source, beam splitter, alliteration photomodulator and radio-frequency (RF) driver circuit, directional trim device, spatial filter, beam separation transducer, condenser lens, microcobjective; Its driving radiofrequency signal is that frequency is identical and have the drive signal of 0～2 π controllable phase difference and a real-time change frequency.

Images