WO2018133589A1

WO2018133589A1 - Aerial photography method, device, and unmanned aerial vehicle

Info

Publication number: WO2018133589A1
Application number: PCT/CN2017/115877
Authority: WO
Inventors: 胡华智
Original assignee: 亿航智能设备（广州）有限公司
Priority date: 2017-01-17
Filing date: 2017-12-13
Publication date: 2018-07-26
Also published as: CN107071389A

Abstract

An aerial photography method, device, and unmanned aerial vehicle. The method comprises the following steps: acquiring images using two cameras; stitching the two images acquired by the two cameras into one 3D image; and transmitting the 3D image. As a result, the unmanned aerial vehicle can provide the 3D image during an aerial photography process, creating real-time lively 3D images for a user to view and bringing immersive experience for the user, significantly enhancing user experience in aerial photography. Moreover, the embodiment can further comprise: detecting a depth using the 3D image while transmitting the 3D image in real time for a user to view. By employs only a set of binocular vision cameras (the two cameras), the unmanned aerial vehicle can implement a plurality of functions such as 3D aerial photography, obstacle avoidance, tracking, and distance detection, without employing two sets of binocular vision cameras (four cameras) for separately implementing 3D aerial photography and depth detection, thereby implementing the plurality of functions at lower cost.

Description

Aerial photography method, device and drone

Technical field

The invention relates to the technical field of drones, in particular to an aerial photography method, a device and a drone.

Background technique

With the rapid development of drone technology, the application of drones has become more and more extensive. Among them, aerial photography is an important application field. The current aerial photography system is mainly composed of a drone and a remote control device. The user controls the drone through a remote control device, and the drone transmits the captured image to the remote control device for viewing by the user through an aerial camera.

In order to provide a more realistic experience for users, in the prior art, a virtual reality (VR) device such as VR glasses is started as a remote control device. The drone transmits the captured image to the VR glasses in real time, and the user watches the captured image in real time through the VR glasses, and the user also controls the posture and shooting angle of the drone through the VR glasses.

Since VR glasses completely isolate the human eye from the outside, it provides an immersive experience for the user and improves the user experience to a certain extent. However, since the image transmitted by the drone to the VR glasses is a 2D image, the advantage of the VR glasses cannot be fully utilized, and the method gives the user an immersive experience.

technical problem

The main object of the embodiments of the present invention is to provide an aerial photography method, device and drone, which aim to realize an immersive experience for the user when performing aerial photography through the drone. .

Technical solution

In order to achieve the above object, an aerial photography method is proposed on the one hand, and the method comprises the following steps:

Acquiring images through two cameras;

Splicing two images acquired by the two cameras into one 3D image;

The 3D image is sent out.

Optionally, the image is a photo or video stream.

Optionally, when the image is a video stream, the splicing the two images collected by the two cameras into one 3D image includes:

The two video streams collected by the two cameras are respectively sampled into two preset resolution video streams, and the preset resolution is lower than the original resolution;

The two preset resolution video streams are spliced into one 3D video stream.

Optionally, the two cameras are arranged side by side, and the splicing the two images collected by the two cameras into one 3D image includes:

The two images collected by the two cameras are spliced side by side to form a 3D image in a left and right format.

Optionally, the two images collected by the two cameras are side-by-side spliced together: the image captured by the left camera is spliced to the left, and the image captured by the right camera is spliced to the right.

Optionally, after the step of splicing the two images collected by the two cameras into one 3D image, the method further includes: performing depth detection of the captured scene by using the 3D image to obtain depth information.

Optionally, the sending the 3D image outward comprises: transmitting the 3D image to a head mounted virtual reality device.

In another aspect, an aerial camera device is provided, the device comprising:

An image acquisition module for collecting images through two cameras;

An image processing module, configured to splicing two images collected by the two cameras into one 3D image;

An image sending module, configured to send the 3D image outward.

Optionally, the image is a photo or video stream.

Optionally, when the image is a video stream, the image processing module is configured to: separately sample two video streams collected by the two cameras into two preset resolution video streams, and the two The preset resolution video streams are spliced into a 3D video stream, wherein the preset resolution is lower than the original resolution.

Optionally, the two cameras are arranged side by side, and the image processing module is configured to: splicing the two images collected by the two cameras side by side to obtain a 3D image in a left and right format.

Optionally, the image processing module is configured to: splicing the image collected by the left camera to the left side, and splicing the image collected by the right camera to the right side.

Optionally, the device further includes a depth detecting module, configured to: perform depth detection of the shooting scene by using the 3D image to obtain depth information.

Optionally, the image sending module is configured to: send the 3D image to the head mounted virtual reality device.

The invention also proposes a drone comprising:

One or more processors;

Memory

One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to use Perform the aforementioned aerial photography method.

Acquiring images through two cameras;

Splicing two images acquired by the two cameras into one 3D image;

The 3D image is sent out.

Optionally, the image is a photo or video stream.

The two preset resolution video streams are spliced into one 3D video stream.

In another aspect, an aerial camera device is provided, the device comprising:

An image acquisition module for collecting images through two cameras;

An image sending module, configured to send the 3D image outward.

Optionally, the image is a photo or video stream.

The invention also proposes a drone comprising:

One or more processors;

Memory

Beneficial effect

An aerial photography method provided by an embodiment of the present invention collects two images through two cameras, and splicing the collected two images into a 3D image and transmitting the same, so that the drone can provide a 3D image during the aerial photography process, and The user can view the realistic 3D images in real time, allowing the user to have an immersive feeling, realizing an immersive experience, and greatly improving the user's aerial photography experience.

Further, while the 3D image is transmitted to the user for viewing in real time, the depth detection is also performed through the 3D image, so that the drone can simultaneously realize 3D aerial photography, obstacle avoidance, and tracking by using a set of binocular cameras (ie, two cameras). A variety of functions, such as ranging, do not need to use two sets of binocular cameras (ie, four cameras) to achieve 3D shooting and depth detection respectively, thereby achieving multiple functions at a lower cost.

DRAWINGS

1 is a flow chart of an aerial photography method according to a first embodiment of the present invention;

2 is a flow chart of an aerial photography method according to a second embodiment of the present invention;

Figure 3 is a block diagram showing an aerial photographing apparatus of a third embodiment of the present invention;

Fig. 4 is a block diagram showing the aerial device of the fourth embodiment of the present invention.

The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments.

Embodiments of the invention

It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

It should be noted that all directional indications (such as up, down, left, right, front, back, ...) in the embodiments of the present invention are only used to explain between components in a certain posture (as shown in the drawing). Relative positional relationship, motion situation, etc., if the specific posture changes, the directional indication also changes accordingly.

In addition, the descriptions of "first", "second", and the like in the present invention are used for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" or "second" may include at least one of the features, either explicitly or implicitly. In addition, the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. It is also within the scope of protection required by the present invention.

The aerial photographing method and the aerial photographing device of the embodiment of the present invention are mainly applied to the drone, and can of course be applied to other aircrafts, which is not limited by the present invention. The embodiment of the present invention is described in detail by applying to an unmanned aerial vehicle as an example.

Embodiment 1

Referring to FIG. 1, an aerial photography method according to a first embodiment of the present invention is proposed. The method includes the following steps:

S11. Acquire an image through two cameras.

In the embodiment of the invention, the drone is provided with two cameras to form a set of binocular cameras. The two cameras are preferably arranged side by side. Of course, it can also be staggered, that is, the two cameras are not on the same horizontal line. The two cameras are separated by a certain distance. In theory, the larger the separation distance, the better.

In this step S11, the drone acquires images simultaneously (synchronously) through two cameras, and the acquired images may be photos or video streams.

S12. Splicing two images acquired by the two cameras into one 3D image.

Specifically, the two images collected by the two cameras are spliced side by side, preferably, the image collected by the left camera is spliced to the left, and the image captured by the right camera is spliced to the right, and finally a 3D (3D) format is obtained. )image. Optionally, the two images collected by the two cameras may be spliced together side by side to obtain a 3D image of the top and bottom format. The 3D image is a 3D photo or a 3D video stream.

Further, before performing image splicing, the drone first performs resolution reduction processing on the original image, and then splicing the reduced resolution image to reduce the final 3D image size, thereby avoiding excessive consumption during subsequent transmission. The bandwidth resources, thereby increasing the transmission speed and improving the real-time performance of image transmission.

Taking the captured image as a video stream as an example, the drone first samples the two video streams collected by the two cameras into two preset resolution video streams, and then splices the two preset resolution video streams into A 3D video stream in which the preset resolution is lower than the original resolution.

For example, the two cameras of the drone respectively capture a video stream of 4K resolution, and the drone samples two video streams of 4K resolution into two video streams of 720P format, and the sampling method can adopt a general downsampling algorithm. For example, the four pixels are merged into one pixel; the sampled two 720P video streams are placed on the left side of the left camera and the right camera is placed on the right side while the frame synchronization is maintained. A 3D video stream with a resolution of 2560*720.

In addition, the drone can save the original images captured by the two cameras in the local storage space. Further, before the saving, the original image is also compressed to save storage space, such as compressing the video stream into a H.265 format video file.

S13. Send a 3D image outward.

In this step S13, the drone transmits the obtained 3D image, for example, to a remote control device or a terminal device that establishes a wireless communication connection with the drone, such as a mobile phone, a tablet computer, or a head mounted virtual reality (VR) device. (such as VR glasses, VR helmets, etc.), etc., or uploaded to a server via a wireless communication network.

Further, before transmitting the 3D image, the drone further performs compression processing on the 3D image to reduce the size of the 3D image, improve transmission efficiency, and realize real-time transmission. For example, a 3D video stream is compressed into a H.264 format video stream and then sent out.

Further, the drone uses 3D images to perform depth detection of the captured scene, acquires depth information, and uses depth information to implement functions such as target ranging, face recognition, gesture recognition, target tracking, and the like, and can be combined with the drone. Attitude information and depth information can be used to avoid obstacles (such as forward obstacle avoidance), so that a set of binocular cameras can simultaneously achieve various functions such as aerial photography, obstacle avoidance, tracking, and ranging.

Wherein, depth detection is performed by using a 3D image, that is, depth detection is performed by using a difference (parallax) between left and right or upper and lower images in a 3D image (such as a 3D video stream). Parallax is the difference in direction produced by observing the same target from two points with a certain distance, so there is parallax in the image obtained by binocular cameras (such as left camera and right camera) of the same target at different positions. The closer the target is to the camera, the larger the parallax in the image of the binocular camera. Therefore, the distance from the target to the camera, that is, the depth of the target, can be calculated according to the parallax size of the target in the two images obtained by the binocular camera. Depth detection. The image is divided into several effective areas, the target distance of each area is calculated in turn, and the distance and the area orientation are fed back to the flight control. The flight control can realize obstacle avoidance according to the distance and orientation of the front target.

Optionally, when the user controls the camera pitch by a remote control device (such as VR glasses) (usually implementing the camera pitch by controlling the pan tilt of the camera), when the pitch angle is greater than the preset pitch angle (can be set according to actual needs) When the drone prompts the user to disable the obstacle function and/or maintain the hover state.

In the embodiment of the present invention, when the user selects a target that needs to be tracked, the drone will adjust the posture of itself and the pan/tilt to align with the selected target. Because the target tracking accuracy based on depth information is more accurate than the previous method of planar vision, it can realize the tracking function with practical application value for the drone.

In the embodiment of the present invention, when the user triggers the photographing instruction, such as waving a wave, a two-hand frame, and the like in front of the remote control device (such as VR glasses), the drone takes two full-resolution photos through two cameras. Store locally and use the two photos you took to stitch into a 3D photo.

Further, the drone can also perform image quality improvement through the respective screens of the two cameras, such as performing denoising processing, background blurring processing, and the like. Specifically, the two photos taken by the binocular camera can be matched by the feature points to find a completely overlapping area, and the picture in the area is equivalent to two shots. Overlay processing (such as the simplest weighted average) after multiple shots of the same picture can effectively reduce noise. At the same time, according to the depth information of the dual purpose, the foreground and the back view of the picture can be distinguished, and the back scene can be blurred by a fuzzy algorithm (such as the simplest Gaussian blur filter), thereby forming a background blur effect.

The aerial photography method of the embodiment of the present invention has the following technical effects as compared with the prior art:

1) Make 3D images by means of image stitching, so that the drone can shoot 3D images through binocular cameras and transmit them to terminal devices such as VR glasses in real time. Users can watch 3D videos in real time through VR glasses, and provide aerial photography in real time. The 3D video source greatly enhances the user experience;

2) Deep detection by 3D image, so that the drone can use the same set of binocular cameras to realize depth detection of the shooting scene, obtain depth information, realize forward obstacle avoidance, target tracking and other functions, without using two sets of double The camera (ie, four cameras) realizes 3D shooting and depth detection respectively, thereby realizing various functions at a low cost;

3) It is also possible to apply the special application based on binocular shooting on the terminal device such as mobile phone to the drone, and further expand the function of the drone;

4) Combining depth information for target tracking improves the accuracy of tracking the subject (such as people and cars), making the drone more reliable and effective for target tracking;

5) Using depth information for target ranging, you can accurately measure the distance of the target, which is safe for close-up shooting.

Embodiment 2

Referring to FIG. 2, an aerial photography method according to a second embodiment of the present invention is proposed. The method includes the following steps:

S21. Collecting a video stream through two cameras arranged side by side.

In this embodiment, the binocular cameras of the drone are arranged side by side, and the drone simultaneously collects the video stream through the binocular camera.

S22. The two video streams collected by the two cameras are respectively sampled into two preset resolution video streams.

S23. The two preset resolution video streams are spliced side by side to obtain a left and right format 3D video stream.

Specifically, the drone firstly samples the two video streams collected by the binocular camera into two preset resolution video streams, and then splicing the two preset resolution video streams side by side, preferably, The video stream collected by the left camera is spliced to the left, and the video stream collected by the right camera is spliced to the right, and finally a 3D video stream of the left and right format is obtained, wherein the preset resolution is lower than the original resolution. For example, the two cameras of the drone each shoot a 4K resolution video stream, and the drone first samples two 4K resolution video streams into two 720P format video streams, and then two 720P format videos. The stream is spliced left and right (or up and down) into a 3D video stream with a resolution of 2560*720 in the left and right format (or top and bottom format).

At the same time, the drone also saves the 4K resolution video stream captured by the two cameras in the local storage space. Further, before the saving, the original video stream is also compressed to save storage space, such as compressing the 4K resolution video stream into a H.265 format video file.

S24. The 3D video stream is compressed and sent to the VR glasses.

In this embodiment, the drone compresses the 3D video stream into a video stream of the H.264 format, and then transmits the video stream to the VR glasses for real-time transmission.

After the VR glasses receive the 3D video stream and play it immediately, the user can watch the 3D video captured by the drone in real time, making the picture more realistic, giving the user an immersive feeling and greatly improving the user experience.

The user can control the flight attitude and shooting angle of the drone through the VR glasses. When the user triggers the camera command, such as waving a wave or a two-hand frame in front of the VR glasses, the drone shoots two through two cameras. Full-resolution photos are stored locally, and the two photos taken are stitched into a 3D photo and sent back to the VR glasses, allowing users to view the captured 3D photos in real time. Further, the drone can also perform image quality improvement through the respective screens of the two cameras, such as performing denoising processing, background blurring processing, and the like.

S25. Perform depth detection of the shooting scene by using the 3D video stream to obtain depth information.

In this embodiment, the drone also uses the 3D video stream to perform depth detection of the shooting scene to acquire depth information. The depth detection is performed by using a 3D video stream, that is, using the difference (parallax) between the left and right video streams in the 3D video stream to implement depth detection.

S26. Perform forward obstacle avoidance, target tracking, and target ranging according to the depth information.

In this embodiment, the depth information is used to implement obstacle avoidance (such as forward obstacle avoidance), face recognition, gesture recognition, target tracking, etc., and the target distance measurement can be realized by combining the attitude information and the depth information of the drone, and the specific implementation is realized. The process is the same as the prior art and will not be described here.

In this embodiment, the two video streams captured by the binocular camera are spliced into one 3D video stream, and the 3D video stream is transmitted to the VR glasses in real time for the user to view in real time, and the depth detection is performed through the 3D video stream, so that the drone utilizes one. The group binocular camera can realize 3D aerial photography, obstacle avoidance, tracking, ranging and other functions at the same time, without using two sets of binocular cameras (ie four cameras) to achieve 3D shooting and depth detection respectively, thus lowering Costs enable multiple functions.

Embodiment 3

Referring to FIG. 3, an aerial device according to a third embodiment of the present invention is proposed. The device includes an image acquisition module, an image processing module, and an image transmission module, wherein:

Image acquisition module: used to capture images through two cameras.

The image acquisition module acquires images simultaneously (synchronously) through two cameras, and the acquired images may be photos or video streams.

Image processing module: used to splicing two images acquired by two cameras into one 3D image.

Specifically, the image processing module splices the two images collected by the two cameras side by side, preferably, the image collected by the left camera is spliced to the left, and the image captured by the right camera is spliced to the right, and finally a left and right format is obtained. 3D (stereo) image. Optionally, the image processing module can also splicing the two images collected by the two cameras side by side, and finally obtaining a 3D image of the top and bottom format. The 3D image is a 3D photo or a 3D video stream.

Further, before performing image splicing, the image processing module first performs resolution reduction processing on the original image, and then splicing the reduced resolution image to reduce the final 3D image size, thereby avoiding excessive consumption during subsequent transmission. The bandwidth resources, thereby increasing the transmission speed and improving the real-time performance of image transmission.

Taking the captured image as a video stream as an example, the image processing module first samples the two video streams collected by the two cameras into two preset resolution video streams, and then splices the two preset resolution video streams into A 3D video stream in which the preset resolution is lower than the original resolution.

For example, the two cameras of the drone respectively capture a video stream of 4K resolution, and the image processing module samples two video streams of 4K resolution into two video streams of 720P format, and the sampling mode can adopt a general downsampling algorithm. For example, combine 4 pixels into one pixel; in the case of keeping the video stream frames of two 720P formats synchronized, the picture taken by the left camera is placed on the left side, and the picture taken by the right camera is placed on the right side, and two 720P are placed. The formatted video stream is spliced left and right into a 3D video stream with a resolution of 2560*720.

In addition, the image processing module can also save the original images captured by the two cameras in a local storage space. Further, before saving, the image processing module further compresses the original image to save storage space, such as compressing the video stream into a video file of the H.265 format.

Image Send Mode: Used to send 3D images outward.

Specifically, the image sending module sends the obtained 3D image in real time (or timing), for example, to a remote control device or a terminal device that establishes a wireless communication connection with the drone, such as a mobile phone, a tablet computer, or a headset virtual reality device. Equipment (such as VR glasses, VR helmets, etc.), etc., or uploaded to a server via a wireless communication network.

Further, before transmitting the 3D image, the image processing module further performs compression processing on the 3D image to reduce the size of the 3D image, improve transmission efficiency, and realize real-time transmission. For example, a 3D video stream is compressed into a H.264 format video stream and then sent out.

In the embodiment of the present invention, when the user triggers a photographing instruction, such as waving a wave, a two-hand frame, or the like in front of a remote control device (such as VR glasses), the image acquisition module takes two full-resolution photos through two cameras. Stored locally, the image processing module uses the two photos taken to form a 3D photo.

Further, the image processing module can also perform image quality improvement through the respective screens of the two cameras, such as performing denoising processing, background blurring processing, and the like. Specifically, the two photos taken by the binocular camera can be completely overlapped by matching the feature points, and the image in the area is equivalent to two shots, and the image processing module performs multiple shots. The same picture is superimposed (such as the simplest weighted average) to effectively reduce noise. At the same time, according to the depth information of the dual purpose, the foreground and the back scene of the picture can be distinguished, and the image processing module can blur the back scene through a fuzzy algorithm (such as the simplest Gaussian blur filtering), thereby forming a background blur effect.

The aerial camera of the embodiment of the invention collects two images through two cameras, and splicing the collected two images into a 3D image and transmitting them, so that the drone can provide 3D images during the aerial photography process, and enables the user to real-time Viewing the realistic 3D image of the screen brings the user an immersive experience, which greatly enhances the user's aerial photography experience.

Embodiment 4

Referring to FIG. 4, an aerial photographing apparatus according to a fourth embodiment of the present invention is proposed. In this embodiment, a depth detecting module is added to the third embodiment, and the depth detecting module is configured to: perform depth detection of a shooting scene by using a 3D image. Obtain depth information, and then use depth information to achieve target ranging, face recognition, gesture recognition, target tracking and other functions, and can be combined with the attitude information and depth information of the drone to achieve obstacle avoidance (such as forward obstacle avoidance). Thus, a set of binocular cameras can simultaneously achieve various functions such as aerial photography, obstacle avoidance, tracking, and ranging.

Wherein, depth detection is performed by using a 3D image, that is, depth detection is performed by using a difference (parallax) between left and right or upper and lower images in a 3D image (such as a 3D video stream). Parallax is the difference in direction produced by observing the same target from two points with a certain distance, so there is parallax in the image obtained by binocular cameras (such as left camera and right camera) of the same target at different positions. The closer the target is to the camera, the larger the parallax in the image of the binocular camera, so the depth detection module can calculate the distance from the target to the camera according to the parallax size of the target in the two images obtained by the binocular camera, ie the depth of the target. To achieve depth detection.

Optionally, when the user controls the camera pitch by a remote control device (such as VR glasses) (usually implementing camera tilt by controlling the pan tilt of the camera), when the pitch angle is greater than the preset pitch angle, the depth detection module prompts The user's obstacle avoidance function is disabled and/or the drone is kept in a hovering state.

In this embodiment, after the two images captured by the binocular camera are spliced into one 3D image, the 3D image is transmitted to the user in real time, and the depth detection is performed through the 3D image, so that the drone uses a set of binocular cameras ( That is to say, two cameras can realize 3D aerial photography, obstacle avoidance, tracking, ranging and other functions at the same time, without using two sets of binocular cameras (ie four cameras) to realize 3D shooting and depth detection respectively, thereby lowering The cost of implementing a variety of functions.

The invention also proposes a drone, the drone comprising: one or more processors; a memory; one or more applications, wherein the one or more applications are stored in the memory and Configured to be performed by the one or more processors, the one or more applications configured to perform an aerial photography method. The aerial photography method includes the steps of: acquiring images by two cameras; splicing two images acquired by the two cameras into one 3D image; and transmitting the 3D images outward. The aerial photography method described in this embodiment is the aerial photography method according to the above embodiment of the present invention, and details are not described herein again.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better. Implementation. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a cell phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present invention.

The preferred embodiments of the present invention have been described above with reference to the drawings, and are not intended to limit the scope of the invention. A person skilled in the art can implement the invention in various variants without departing from the scope and spirit of the invention. For example, the features of one embodiment can be used in another embodiment to obtain a further embodiment. Any modifications, equivalent substitutions and improvements made within the technical concept of the present invention should be used herein.

Industrial applicability

Further, while the 3D image is transmitted to the user for viewing in real time, the depth detection is also performed through the 3D image, so that the drone can simultaneously realize 3D aerial photography, obstacle avoidance, and tracking by using a set of binocular cameras (ie, two cameras). A variety of functions, such as ranging, do not need to separately use two sets of binocular cameras (ie, four cameras) to achieve 3D shooting and depth detection, respectively, thereby achieving a variety of functions at a lower cost, and therefore, industrial applicability.

Claims

An aerial photography method comprising the following steps:

Acquiring images through two cameras;

Splicing two images acquired by the two cameras into one 3D image;

The 3D image is sent out.
The aerial photography method of claim 1 wherein the image is a photo or video stream.
The aerial photography method according to claim 2, wherein when the image is a video stream, the splicing the two images collected by the two cameras into one 3D image comprises:

The two video streams collected by the two cameras are respectively sampled into two preset resolution video streams, and the preset resolution is lower than the original resolution;

The two preset resolution video streams are spliced into one 3D video stream.
The aerial photography method according to claim 1, wherein the two cameras are arranged side by side, and the splicing the two images collected by the two cameras into one 3D image comprises:

The two images collected by the two cameras are spliced side by side to form a 3D image in a left and right format.
The aerial photography method according to claim 4, wherein said splicing the two images collected by the two cameras side by side side by side comprises:

The image captured by the left camera is stitched to the left, and the image captured by the right camera is stitched to the right.
The aerial photography method according to any one of claims 1-5, wherein the step of splicing the two images acquired by the two cameras into one 3D image further comprises:

The depth detection of the captured scene is performed by using the 3D image to obtain depth information.
The aerial photography method according to any one of claims 1 to 5, wherein the outward transmission of the 3D image comprises transmitting the 3D image to a head mounted virtual reality device.
An aerial camera device comprising:

An image acquisition module for collecting images through two cameras;

An image processing module, configured to splicing two images collected by the two cameras into one 3D image;

An image sending module, configured to send the 3D image outward.
The aerial camera of claim 8 wherein the image is a photo or video stream.
The aerial image device according to claim 9, wherein when the image is a video stream, the image processing module is configured to separately sample two video streams collected by the two cameras into two preset resolutions. The video stream is spliced into two 3D video streams, wherein the preset resolution is lower than the original resolution.
The aerial photographing device according to claim 8, wherein the two cameras are arranged side by side, the image processing module is configured to: splicing the two images collected by the two cameras side by side to obtain a left and right format. 3D image.
The aerial image capturing device according to claim 11, wherein the image processing module is configured to: splicing the image captured by the left camera to the left and stitching the image captured by the right camera to the right.
The aerial photographing apparatus according to any one of claims 8 to 12, wherein the apparatus further comprises a depth detecting module, wherein the depth detecting module is configured to: perform depth detection of the shooting scene by using the 3D image to acquire depth information.
The aerial photographing apparatus according to any one of claims 8 to 12, wherein the image transmitting module is configured to: transmit the 3D image to a head mounted virtual reality device.
A drone that includes:

One or more processors;

Memory

One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to use The method of any one of claims 1 to 7 is performed.