Depth perception system and method based on multi-plane rapid scanning

Abstract

The invention provides a depth perception system and method based on multi-plane fast scanning, and the system comprises an image collection assembly which is used for obtaining a two-dimensional scene image of a target three-dimensional scene; the axial scanning device is used for forming a divergent beam according to the divergent image of the two-dimensional scene image and determining axial scanning images of the target three-dimensional scene at different axial positions according to the divergent beam; the multi-plane imaging device is used for mapping the axial scanning images at different axial positions to different transverse positions of the same axial position to obtain a transverse scanning image sequence; and the calculation component is used for calculating perception depth information of the target three-dimensional scene according to the transverse scanning image sequence. The system not only realizes high-precision depth perception of each axial position of the target three-dimensional scene, but also improves the axial scanning speed and the three-dimensional depth perception rate.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional depth perception, and particularly to a depth perception system and method based on multi-plane rapid scanning. Background Art

[0002] With the development of three-dimensional depth perception technology, image-based three-dimensional depth perception has become an indispensable technology in multiple fields such as unmanned aerial vehicle navigation, virtual reality / augmented reality (VR / AR), and autonomous driving. This technology mainly collects two-dimensional or two-dimensional images of multiple perspectives of the target object through optical methods and devices, and uses these images to obtain the depth information of the target object to achieve the reconstruction of three-dimensional information. Traditionally, lidar based on time-of-flight technology is one of the important means to achieve this goal, which determines the depth information of the object by calculating the time for light to travel to and from the target object. However, due to the limitations of the single-point acquisition method, lidar faces challenges in large field of view, high resolution, and high-speed acquisition, and it is difficult to meet the above requirements simultaneously.

[0003] In order to overcome the limitations of lidar, image-based depth perception technology has gradually become a research hotspot. This type of technology collects two-dimensional images of the scene through a single exposure of an area array camera, and uses prior information to restore the depth information of the scene, thereby greatly improving the spatio-temporal bandwidth product. Currently, depth perception methods based on multi-view stereo matching technology and light field technology are widely used, but they each have obvious defects. For example, the multi-view stereo matching technology performs poorly in dealing with complex occlusion scenes, while the light field technology faces the problem of mutual restriction between spatial resolution and angular resolution. In addition, although the monocular depth perception technology based on aberration solves the problem of reduced spatial resolution of the light field technology, it still has deficiencies in dead zones and perception accuracy.

[0004] Furthermore, the monocular depth perception technology based on defocus (Depth from Defocus, DFD) and the depth perception technology based on focus (Focus-based Depth Sensing, FFD) respectively attempt to solve the above problems, but they each also face challenges. The DFD technology is limited by the physical limitations of the system, there is a contradiction between depth perception accuracy and range, and the perception accuracy is extremely low near the focal plane; although the FFD technology improves the depth perception accuracy, due to the need for layer-by-layer scanning, the imaging rate is low, and it cannot meet the requirements of high-speed large field of view three-dimensional information acquisition.

[0005] Therefore, it is of great significance to develop a new type of three-dimensional depth perception technology that can simultaneously take into account high precision, large field of view, and fast acquisition capabilities. Summary of the Invention

[0006] The present invention provides a depth perception system and method based on multi-plane fast scanning, which are used to overcome the defects of low perception accuracy, low three-dimensional imaging rate, and low resolution in the existing three-dimensional depth perception technology, and to achieve high-speed three-dimensional imaging while ensuring the resolution, and to improve the three-dimensional depth perception rate of object imaging.

[0007] On the one hand, the present invention provides a depth perception system based on multi-plane fast scanning, including: an image acquisition component, which is used to obtain a two-dimensional scene image of a target three-dimensional scene; an axial scanning device, which is used to form a divergent light beam according to the divergent image of the two-dimensional scene image, and to determine the axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam; a multi-plane imaging device, which is used to map the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; and a calculation component, which is used to calculate the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images.

[0008] Further, the image acquisition component includes: a collection lens, which is used to perform two-dimensional imaging on the target three-dimensional scene to obtain the two-dimensional scene image.

[0009] Further, the axial scanning device includes: a first lens, which is used to collect the divergent image of the two-dimensional scene image and integrate it to form the divergent light beam according to the divergent image; an electro-optically tunable lens, which is used to converge or diverge the divergent light beam to obtain an adjusted light beam; and a second lens, which is used to focus the adjusted light beam to obtain the axial scanning images of the target three-dimensional scene at different axial positions.

[0010] Further, the multi-plane imaging device includes: a beam splitting prism, which is used to receive the adjusted light beam after being focused by the second lens, and to map the axial scanning images at different axial positions carried in the adjusted light beam to different lateral positions at the same axial position to obtain an outgoing light beam in the initial dimension; a transmissive grating, which is used to disperse the outgoing light beam in the initial dimension to the target dimension to obtain target light beams with different spectra; and a camera, which is used to image the target light beams to obtain the sequence of lateral scanning images.

[0011] Further, the calculation component is specifically used to reconstruct the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images, their corresponding different axial positions, and the system prior information.

[0012] Further, the system prior information includes one or a combination of more of the focusing position of the multi-plane imaging device, the oscillatory change of the axial scanning device over time, and the depth change caused by chromatic aberration.

[0013] Second aspect, the present invention further provides a depth perception method based on multi-plane fast scanning, which is applied to the depth perception system based on multi-plane fast scanning as described in any one of the above, and includes: obtaining a two-dimensional scene image of a target three-dimensional scene; forming a diverging light beam according to the diverging image of the two-dimensional scene image, and determining axial scanning images of the target three-dimensional scene at different axial positions according to the diverging light beam; mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; and calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images.

[0014] Further, the forming a diverging light beam according to the diverging image of the two-dimensional scene image, and determining axial scanning images of the target three-dimensional scene at different axial positions according to the diverging light beam includes: collecting the diverging image of the two-dimensional scene image, and integrally forming the diverging light beam according to the diverging image; converging or diverging the diverging light beam to obtain an adjusted light beam; and focusing the adjusted light beam to obtain the axial scanning images of the target three-dimensional scene at different axial positions.

[0015] Further, the mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images includes: receiving the adjusted light beam passing through the second lens, and mapping the axial scanning images at different axial positions carried in the adjusted light beam to different lateral positions at the same axial position to obtain an outgoing light beam of an initial dimension; dispersing the outgoing light beam of the initial dimension to a target dimension to obtain target light beams of different spectra; and imaging the target light beams to obtain the sequence of lateral scanning images.

[0016] Further, the calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images includes: reconstructing the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images and their corresponding different axial positions, and system prior information; wherein the system prior information includes one or more combinations of the focusing position of the multi-plane imaging device, the oscillating change of the axial scanning device over time, and the depth change caused by chromatic aberration.

[0017] The depth perception system based on multi-plane fast scanning provided by the present invention includes: an image acquisition component for acquiring a two-dimensional scene image of a target three-dimensional scene; an axial scanning device for forming a divergent light beam according to the divergent image of the two-dimensional scene image and determining axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam; a multi-plane imaging device for mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; and a calculation component for calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images. The system scans the axial scanning images of the target three-dimensional scene at different axial positions through a high-speed axial scanning device, and maps the axial scanning images at different axial positions to different regions in a certain dimension of the camera through a multi-plane imaging device, which not only realizes high-precision depth perception of each axial position of the target three-dimensional scene, but also improves the axial scanning speed and the three-dimensional depth perception rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 FIG. is a system schematic diagram of the depth perception system based on multi-plane fast scanning provided by an embodiment of the present invention.

[0020] Figure 2 FIG. is an overall system schematic diagram of the depth perception system based on multi-plane fast scanning provided by an embodiment of the present invention.

[0021] Figure 3 FIG. is a principle schematic diagram of the depth perception system based on multi-plane fast scanning provided by an embodiment of the present invention.

[0022] Figure 4 FIG. is a flow schematic diagram of the depth perception method based on multi-plane fast scanning provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0024] It should be noted that the existing DFD technology is limited by the physical limitations of the system, there is a contradiction between the depth perception accuracy and the range, and the perception accuracy is extremely low near the focal plane. Although the existing FFD technology improves the depth perception accuracy, due to the need for layer-by-layer scanning, the imaging rate is low and cannot meet the requirements of high-speed large-field-of-view three-dimensional information acquisition.

[0025] Considering this, the present invention proposes a new depth perception system, namely, a depth perception system based on multi-plane fast scanning. Specifically, Figure 1 Fig. shows a system schematic diagram of the depth perception system based on multi-plane fast scanning provided by an embodiment of the present invention.

[0026] As Figure 1 shown, the system includes: an image acquisition component 110, configured to acquire a two-dimensional scene image of a target three-dimensional scene; an axial scanning device 120, configured to form a divergent light beam according to the divergent image of the two-dimensional scene image, and determine an axial scanning image of the target three-dimensional scene at different axial positions according to the divergent light beam; a multi-plane imaging device 130, configured to map the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; and a calculation component 140, configured to calculate the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images.

[0027] The following will elaborate on each module of the depth perception system based on multi-plane fast scanning.

[0028] Regarding the image acquisition component.

[0029] In this embodiment, the image acquisition component is mainly used for two-dimensional imaging of a target three-dimensional scene. The image acquisition component may specifically adopt a large-aperture acquisition lens. A large-aperture acquisition lens refers to a lens with a relatively large aperture diameter, which can allow more light to enter the lens, thereby improving the brightness and clarity of the acquired image.

[0030] It is easy to understand that the light reflected or emitted by the object in the target three-dimensional scene is collected by the large-aperture objective lens and will form a primary image plane near the focal position behind the lens. The image formed on the primary image plane is the primary projection of the target three-dimensional scene, that is, the two-dimensional scene image of the target three-dimensional scene. Since the depth of field of the two-dimensional scene image acquired in this embodiment is small, high-perception depth information at a single axial position can be obtained according to the two-dimensional scene image.

[0031] In a specific embodiment, the image acquisition component includes an acquisition lens, configured to perform two-dimensional imaging on a target three-dimensional scene to obtain a two-dimensional scene image. Among them, the role of the acquisition lens is to capture the light from the target three-dimensional scene and focus it on an imaging plane, thereby obtaining a two-dimensional scene image.

[0032] Regarding the axial scanning device.

[0033] The axial scanning device is a device used to achieve precise movement or adjustment along the optical axis direction in an optical system. Common axial scanning devices include, for example, piezoelectric actuators, stepper motors, voice coil motors, and mechanical stages, etc. The main functions of the axial scanning device include: (1) Precise displacement control: providing high-precision movement along the optical axis direction to ensure that the focal position of the optical system can be accurately adjusted; (2) Auto-focusing to keep the image always clear; (3) Depth scanning: capable of constructing a three-dimensional image by scanning different depth planes of the sample layer by layer.

[0034] In this embodiment, the axial scanning device is mainly used to perform high-speed scanning on different axial positions of the target three-dimensional scene to achieve high-precision depth perception of each axial position of the target three-dimensional scene.

[0035] In a specific embodiment, the axial scanning device includes: a first lens for collecting the divergent image of the two-dimensional scene image and integrating it to form a divergent light beam according to the divergent image; an electrically tunable lens for converging or diverging the divergent light beam to obtain an adjusted light beam; a second lens for focusing the adjusted light beam to obtain an axial scanning image of the target three-dimensional scene at different axial positions.

[0036] It is easy to understand that the focusing of the target three-dimensional scene at the primary image plane will diverge again, be collected by the first lens and integrated to form approximately parallel light (i.e., the divergent light beam), and then the parallel approximate light propagates into the electrically tunable lens located at the focal length of the first lens. The electrically tunable lens performs a converging or diverging action on the incoming divergent light beam to obtain an adjusted light beam. Finally, the adjusted light beam is focused by the second lens to obtain an axial scanning image of the target three-dimensional scene at different axial positions, realizing the axial scanning of the target three-dimensional scene.

[0037] Specifically, the first lens in this embodiment has two functions: (1) Beam collection: collecting the light beam diverging from the focus of the primary image plane; (2) Collimation: converting the collected light beam into approximately parallel light, denoted as the divergent light beam. The first lens is usually a convex lens, such as a biconvex lens or a plano-convex lens, and a suitable focal length is selected according to the divergence angle of the light beam.

[0038] The electrically tunable lens dynamically adjusts its focal length through an electrical signal to achieve divergence or focusing of the divergent light beam. The electrically tunable lens can specifically be a liquid lens based on the electrowetting effect, piezoelectric drive, or liquid crystal technology.

[0039] The second lens further optimizes the shape or imaging quality of the adjusted light beam to ensure that the final axial scanning image is clear and meets the expectations. The second lens can be a convex lens, a compound lens, or a combination of multiple lenses.

[0040] Regarding a multi-plane imaging device.

[0041] In this embodiment, the multi-plane imaging device is mainly used to map image planes (axial scan images) at different axial positions to different regions in a certain dimension on the camera.

[0042] In a specific embodiment, the multi-plane imaging device includes: a beam-splitting prism, configured to receive the beam after being adjusted by the second lens, and map the axial scan images at different axial positions carried in the adjusted beam to different lateral positions at the same axial position, to obtain an output beam in the initial dimension; a transmission grating, configured to disperse the output beam in the initial dimension to the target dimension, to obtain target beams with different spectra; and a camera, configured to image the target beams to obtain a sequence of lateral scan images.

[0043] It is easy to understand that the beam after being adjusted by the second lens passes through the beam-splitting prism before focusing. The beam-splitting prism maps scenes (axial scan images) at different axial positions to different lateral positions at the same axial position, and the beams at the exits of the beam-splitting prism, that is, the output beam in the initial dimension, can be obtained. Subsequently, the output beam in the initial dimension passes through the transmission grating. The transmission grating divides the output beam into target beams with different spectra, and finally images on the camera to obtain a sequence of lateral scan images.

[0044] Among them, the beam-splitting prism controls the path of light through a specific geometric shape and optical coating, and divides an incident beam (the adjusted beam) into two or more beams (the output beam in the initial dimension). Common beam-splitting prisms include cube beam-splitting prisms, flat beam splitters, and polarization beam splitters.

[0045] The transmission grating is used to decompose an incident beam (the output beam in the initial dimension) into its corresponding wavelength components. The transmission grating consists of a series of parallel slits or rulings. These slits or rulings can be opaque lines on a transparent material, or opaque regions between transparent lines. When a beam passes through these slits or rulings, a diffraction phenomenon occurs, generating multiple diffraction orders, thereby separating light with different wavelengths. Common transmission gratings include regular gratings, blazed gratings, and holographic gratings.

[0046] The initial dimension is the same as the number of exits of the beam-splitting prism, and the initial dimension is the same as the number of axial positions of the scanned target three-dimensional scene. The target dimension can be set according to actual situations and is not specifically limited herein. For example, in one embodiment, the initial dimension is 3×1 and the target dimension is 3×3.

[0047] It should be noted that the multi-plane imaging device in this embodiment further discretizes the area between two different axial positions by mapping images of different wavelengths to different camera areas in another dimension, reducing the scanning range and further improving the three-dimensional acquisition rate.

[0048] Regarding the computing component.

[0049] In this embodiment, the computing component is specifically configured to reconstruct the perceived depth information of the target three-dimensional scene based on the lateral scanning image sequence and its corresponding different axial positions, as well as the system prior information. Among them, the system prior information includes one or a combination of the focusing position of the multi-plane imaging device, the oscillatory change of the axial scanning device over time, and the depth change caused by chromatic aberration.

[0050] It is easy to understand that the computing component matches the lateral scanning images collected by the camera at each moment with the depth position (axial position) focused by the system according to the prior information such as the focusing position of the multi-plane imaging device, the oscillatory change of the axial scanning device over time, and the depth change caused by chromatic aberration, and finally obtains the perceived depth information of the target three-dimensional scene.

[0051] Among them, the focusing position of the multi-plane imaging device is a specific distance point at which the image can reach the best clarity during the process of the camera imaging the target light beam. The oscillatory change of the axial scanning device over time refers to the periodic vibration of the focal length of the electrically tunable lens over time.

[0052] Chromatic aberration is a common aberration phenomenon in optical systems, which occurs when light of different wavelengths passes through a lens and cannot be focused at the same point due to different refractive indices. The chromatic aberration in this embodiment especially refers to longitudinal chromatic aberration, which means that light of different wavelengths cannot be focused on the same plane on the same axis, but is distributed at different distances. In the multi-plane imaging device, this may cause information of different colors to come from different depths, thus affecting the overall three-dimensional reconstruction accuracy.

[0053] The perceived depth information of the target three-dimensional scene includes but is not limited to depth maps, point cloud data, and texture information. The depth map includes the depth value corresponding to each pixel point in the lateral scanning image, that is, the distance from each pixel point to the camera. The point cloud data can be obtained by converting each pixel in the depth map into its corresponding three-dimensional coordinates. The texture information refers to the color or gray value of each pixel point in the lateral scanning image.

[0054] In this embodiment, the depth perception system based on multi-plane fast scanning includes: an image acquisition component for acquiring a two-dimensional scene image of a target three-dimensional scene; an axial scanning device for forming a divergent light beam according to the divergent image of the two-dimensional scene image and determining the axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam; a multi-plane imaging device for mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; and a calculation component for calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images. This system scans to obtain the axial scanning images of the target three-dimensional scene at different axial positions through a high-speed axial scanning device, and maps the axial scanning images at different axial positions to different regions in a certain dimension of the camera through a multi-plane imaging device, which not only realizes high-precision depth perception of each axial position of the target three-dimensional scene, but also improves the axial scanning speed and the three-dimensional depth perception rate.

[0055] In some other embodiments, Figure 2 Fig. shows the overall system schematic diagram of the depth perception system based on multi-plane fast scanning provided by the embodiments of the present invention.

[0056] As Figure 2 shown, first, the light beam reflected or emitted by the object in the target three-dimensional scene 201 passes through the large-aperture acquisition lens 202, and a primary image plane 203 is formed near the rear focal plane of the acquisition lens. The image presented on the primary image plane is the two-dimensional scene image.

[0057] The light beam emitted or reflected by the object in the target three-dimensional scene diverges again on the primary image plane 203, is collected and integrated by the first lens 204 to form an approximately parallel light beam (divergent light beam), and then propagates into the electro-optic lens 205 located at the focal length of the first lens 204. The electro-optic lens 205 converges or diverges the incoming approximately parallel light beam to obtain an adjusted light beam. Finally, the adjusted light beam is focused by the second lens 206 to obtain the axial scanning images of the target three-dimensional scene at different axial positions, thereby realizing the axial scanning of the target three-dimensional scene.

[0058] Before the adjusted light beam passing through the second lens 206 is about to be focused, it passes through the beam splitting prism 207, maps the axial scanning images at different axial positions to different lateral positions at the same axial position, obtains the outgoing light beams at the exits of the beam splitting prism, and then the outgoing light beams pass through the transmissive grating 208 to divide the outgoing light beams into target light beams of different spectra, and finally are imaged on the camera 209 to obtain a sequence of lateral scanning images.

[0059] Furthermore, Figure 3 Fig. shows the principle schematic diagram of the depth perception system based on multi-plane fast scanning provided by the embodiments of the present invention. As Figure 3As shown, different planes 1, 2, 3, 4 (i.e., axial scan images 1, 2, 3, 4 at different axial positions) are mapped to different lateral positions at the same axial position through different optical paths of the beam-splitting prism 207. Each plane is then dispersed to another dimension (the target dimension) through the transmissive grating 208. Taking plane 1 as an example, four images 1.1, 1.2, 1.3, 1.4 with different wavelengths are formed after dispersion and mapped to different positions of the camera, thereby obtaining a sequence of lateral scan images.

[0060] Corresponding to the depth perception system based on multi-plane fast scanning described in the above embodiments, the present invention also proposes a depth perception method based on multi-plane fast scanning. Specifically, Figure 4 FIG. shows a schematic flowchart of the depth perception method based on multi-plane fast scanning provided by the embodiments of the present invention.

[0061] As Figure 4 shown, the method includes: S410, obtaining a two-dimensional scene image of the target three-dimensional scene; S420, forming a divergent light beam according to the divergent image of the two-dimensional scene image, and determining axial scan images of the target three-dimensional scene at different axial positions according to the divergent light beam; S430, mapping axial scan images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scan images; S440, calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scan images.

[0062] In a specific embodiment, forming a divergent light beam according to the divergent image of the two-dimensional scene image and determining axial scan images of the target three-dimensional scene at different axial positions according to the divergent light beam includes: collecting the divergent image of the two-dimensional scene image and integrating it to form a divergent light beam; converging or diverging the divergent light beam to obtain an adjusted light beam; focusing the adjusted light beam to obtain axial scan images of the target three-dimensional scene at different axial positions.

[0063] In another specific embodiment, mapping axial scan images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scan images includes: receiving the adjusted light beam passing through the second lens, and mapping axial scan images at different axial positions carried in the adjusted light beam to different lateral positions at the same axial position to obtain an outgoing light beam in the initial dimension; dispersing the outgoing light beam in the initial dimension to the target dimension to obtain target light beams with different spectra; imaging the target light beams to obtain a sequence of lateral scan images.

[0064] In still another specific embodiment, the perceived depth information of the target three-dimensional scene is calculated according to the lateral scan image sequence, including: reconstructing the perceived depth information of the target three-dimensional scene according to the lateral scan image sequence, its corresponding different axial positions, and the system prior information; wherein the system prior information includes one or a combination of the focusing position of the multi-plane imaging device, the oscillatory change of the axial scanning device over time, and the depth change caused by chromatic aberration.

[0065] It should be noted that the depth perception method based on multi-plane fast scanning provided by the embodiments of the present invention can be correspondingly referred to the depth perception device based on multi-plane fast scanning described in the above embodiments, and will not be elaborated here.

[0066] In this embodiment, by obtaining the two-dimensional scene image of the target three-dimensional scene, a divergent light beam is formed according to the divergent image of the two-dimensional scene image, and the axial scan images of the target three-dimensional scene at different axial positions are determined according to the divergent light beam. Furthermore, the axial scan images at different axial positions are mapped to different lateral positions at the same axial position to obtain a lateral scan image sequence. Thus, according to the lateral scan image sequence, the perceived depth information of the target three-dimensional scene is calculated. This method scans the axial scan images of the target three-dimensional scene at different axial positions through a high-speed axial scanning device, and maps the axial scan images at different axial positions to different regions in a certain dimension of the camera through a multi-plane imaging device, which not only realizes high-precision depth perception of each axial position of the target three-dimensional scene, but also improves the axial scanning speed and the three-dimensional depth perception rate.

[0067] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A depth perception system based on multi-plane fast scanning, characterized in that, Comprising: An image acquisition component, configured to acquire a two-dimensional scene image of a target three-dimensional scene; An axial scanning device, configured to form a divergent light beam according to the divergent image of the two-dimensional scene image, and determine axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam; A multi-plane imaging device, configured to map the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; A calculation component, configured to calculate the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images.

2. The depth perception system based on multi-plane fast scanning according to claim 1, wherein, The image acquisition component includes: An acquisition lens, configured to perform two-dimensional imaging on the target three-dimensional scene to obtain the two-dimensional scene image.

3. The depth perception system based on multi-plane fast scanning according to claim 1, wherein The axial scanning device includes: A first lens, configured to collect the divergent image of the two-dimensional scene image and integrally form the divergent light beam according to the divergent image; An electro-tunable lens, configured to converge or diverge the divergent light beam to obtain an adjusted light beam; A second lens, configured to focus the adjusted light beam to obtain axial scanning images of the target three-dimensional scene at different axial positions.

4. The depth perception system based on multi-plane fast scanning according to claim 3, characterized in that, The multi-plane imaging device includes: A beam splitting prism, configured to receive the adjusted light beam after being focused by the second lens, and map the axial scanning images at different axial positions carried in the adjusted light beam to different lateral positions at the same axial position to obtain an output light beam of an initial dimension; A transmissive grating, configured to disperse the output light beam of the initial dimension to a target dimension to obtain target light beams of different spectra; A camera, configured to image the target light beams to obtain the sequence of lateral scanning images.

5. The depth perception system based on multi-plane fast scanning according to claim 1, wherein The calculation component is specifically configured to reconstruct and obtain the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images, their corresponding different axial positions, and system prior information.

6. The depth perception system based on multi-plane fast scanning according to claim 5, characterized in that, The system prior information includes one or a combination of multiple items such as the focusing position of the multi-plane imaging device, the oscillating change of the axial scanning device over time, and the depth change caused by chromatic aberration.

7. A depth perception method based on multi-plane fast scanning, applied to the depth perception system based on multi-plane fast scanning according to any one of claims 1-6, characterized in that, Comprising: Acquiring a two-dimensional scene image of a target three-dimensional scene; Forming a divergent light beam according to the divergent image of the two-dimensional scene image, and determining axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam; Mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images; Calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images.

8. The depth perception method based on multi-plane fast scanning according to claim 7, characterized in that The forming a divergent light beam according to the divergent image of the two-dimensional scene image, and determining axial scanning images of the target three-dimensional scene at different axial positions according to the divergent light beam includes: Collecting the divergent image of the two-dimensional scene image and integrally forming the divergent light beam according to the divergent image; Converging or diverging the divergent light beam to obtain an adjusted light beam; Focusing the adjusted light beam to obtain axial scanning images of the target three-dimensional scene at different axial positions.

9. The depth perception method based on multi-plane fast scanning according to claim 7, characterized in that, The mapping the axial scanning images at different axial positions to different lateral positions at the same axial position to obtain a sequence of lateral scanning images includes: Receive the light beam after being focused by the second lens, and map the axial scanning images at different axial positions carried in the focused light beam to different lateral positions at the same axial position, so as to obtain an outgoing light beam with an initial dimension; Disperse the outgoing light beam with the initial dimension to a target dimension to obtain target light beams with different spectra; Image the target light beams to obtain the sequence of lateral scanning images.

10. The depth perception method based on multi-plane fast scanning according to claim 7, characterized in that, The calculating the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images includes: Reconstruct the perceived depth information of the target three-dimensional scene according to the sequence of lateral scanning images, their corresponding different axial positions, and the system prior information; Wherein, the system prior information includes one or a combination of more of the focusing position of the multi-plane imaging device, the oscillating change of the axial scanning device over time, and the depth change caused by chromatic aberration.

Images