CN117716702A

CN117716702A - Image shooting method and device and movable platform

Info

Publication number: CN117716702A
Application number: CN202280050536.7A
Authority: CN
Inventors: 刘宝恩; 王涛; 李鑫超
Original assignee: SZ DJI Technology Co Ltd
Current assignee: SZ DJI Technology Co Ltd
Priority date: 2022-02-28
Filing date: 2022-02-28
Publication date: 2024-03-15
Also published as: WO2023159611A1

Abstract

An image shooting method, an image shooting device and a movable platform. The method comprises the following steps: collecting a first target image at a first preset focal length, and identifying a target object in the first target image; when the target object is identified in the first target image, adjusting the position and the posture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the shooting device; when the first target image cannot identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length; in the process of adjusting the focal length of the image pickup device, when a target object is identified according to the current image acquired by the image pickup device at any moment, the position and the gesture of the movable platform are adjusted according to the position of the target object in the current image. According to the embodiment of the disclosure, the accuracy of unmanned automatic shooting can be improved.

Description

Image shooting method and device and movable platform

Technical Field

The disclosure relates to the technical field of unmanned control, in particular to an image shooting method and device capable of improving unmanned automatic shooting accuracy and a movable platform.

Background

With the development of unmanned aerial vehicle technology, tasks such as electric power inspection, bridge inspection, oil and gas pipeline inspection and the like, which need to repeatedly inspect targets, are gradually born by unmanned aerial vehicles. In the inspection process, the unmanned aerial vehicle needs to accurately photograph/record the target so as to compare and check the state of the inspected target in the later period to check the working state of the key equipment. In the related art, there is a zoom lens teaching/replay-based inspection scheme, that is, the position, the gesture and the focal length of a movable platform for shooting a target object in the teaching process are recorded, and the target object is automatically shot according to the position, the gesture and the focal length in the replay process. However, when the target object is positioned in the replay process, the time for adjusting the position, the posture and the focal length of the cradle head is long, and shooting errors caused by cradle head control deviation in the teaching process, body shaking caused by strong wind in the actual shooting process and the like cannot be solved.

It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the present disclosure and thus may include information that does not constitute prior art known to those of ordinary skill in the art.

Disclosure of Invention

The disclosure aims to provide an image shooting method, an image shooting device and a movable platform, which are used for solving the problem of inaccurate target object positioning in an unmanned shooting process in the related technology at least to a certain extent.

According to a first aspect of embodiments of the present disclosure, there is provided an image capturing method applied to a movable platform, on which an image capturing device is mounted, the movable platform capturing an image by the image capturing device, the method including: collecting a first target image at a first preset focal length, and identifying a target object in the first target image; when the target object is identified in the first target image, adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the camera; when the first target image does not identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length; and in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so as to enable the position of the target object to be located at a preset position in a picture shot by the image pickup device.

According to a second aspect of the present disclosure, there is provided an image capturing apparatus including: a memory configured to store program code; one or more processors coupled to the memory, the processors configured to perform the following methods based on instructions stored in the memory: collecting a first target image at a first preset focal length, and identifying a target object in the first target image; when the target object is identified in the first target image, adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the camera; when the first target image does not identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length; and in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so that the position of the target object is positioned at a preset position in a picture shot by the image pickup device.

According to a third aspect of the present disclosure there is provided a moveable platform comprising: a body; the power system is arranged on the machine body and is used for providing power for the movable platform; the camera device is arranged on the machine body and used for collecting images; a memory; and a processor coupled to the memory, the processor configured to perform the image capture method of any of the above based on instructions stored in the memory.

According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon a program which, when executed by a processor, implements the image capturing method according to any one of the above.

According to the embodiment of the disclosure, the target object is positioned at a shorter focal length, the pose of the movable platform is continuously adjusted, the focal length is increased, the target object is repositioned, and the alignment of the target object can be continuously realized through an iterative alignment process, so that when the unmanned movable platform automatically shoots, the movable platform overcomes the problems of inaccurate given position (control error in the teaching process), body shaking caused by strong wind and other external forces, and the movable platform can accurately realize unmanned automatic shooting.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure. It will be apparent to those of ordinary skill in the art that the drawings in the following description are merely examples of the disclosure and that other drawings may be derived from them without undue effort.

Fig. 1 is a flowchart of an image photographing method in an exemplary embodiment of the present disclosure.

Fig. 2 is a sub-flowchart of step S102 in one embodiment of the present disclosure.

Fig. 3 is a sub-flowchart of step S104 in one embodiment of the present disclosure.

Fig. 4 is a sub-flowchart of step S106 in one embodiment of the present disclosure.

Fig. 5 is a flowchart of an image capturing method in one embodiment of the present disclosure.

Fig. 6 is a flowchart of an image photographing method in another embodiment of the present disclosure.

Fig. 7 is a schematic diagram of a movable platform in one embodiment of the present disclosure.

Fig. 8 is a block diagram of an image photographing apparatus in an exemplary embodiment of the present disclosure.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the present disclosure. One skilled in the relevant art will recognize, however, that the aspects of the disclosure may be practiced without one or more of the specific details, or with other methods, components, devices, steps, etc. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present disclosure.

Furthermore, the drawings are only schematic illustrations of the present disclosure, in which the same reference numerals denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor devices and/or microcontroller devices.

The following describes example embodiments of the present disclosure in detail with reference to the accompanying drawings.

Fig. 1 is a flowchart of an image photographing method in an exemplary embodiment of the present disclosure. The method shown in fig. 1 can be applied to a movable platform, wherein an image pickup device is mounted on the movable platform, and the movable platform collects images through the image pickup device.

Referring to fig. 1, an image photographing method 100 may include:

step S102, a first target image is acquired at a first preset focal length, and a target object is identified in the first target image;

step S104, when the target object is identified in the first target image, adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so that the position of the target object is positioned at a preset position in a shooting picture of the shooting device;

step S106, when the first target image does not identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image collected by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length; and in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so that the position of the target object is positioned at a preset position in a picture shot by the image pickup device.

In the embodiment of the present disclosure, the movable platform may be, for example, an unmanned aerial vehicle (unmanned aerial vehicle) on which the image pickup device is mounted. In one embodiment, the movable platform further comprises a cradle head mounted on the unmanned aerial vehicle, and the camera device is mounted on the cradle head.

The method provided by the embodiment of the disclosure can be applied to a replay stage of a target detection task and a tracking stage of a target tracking task so as to automatically focus and automatically shoot a target object.

The replay stage of the target detection task refers to that a movable platform (such as an unmanned aerial vehicle) is manually controlled in the teaching stage of the target detection task, the shooting position, the shooting height, the focal length and the like of a target to be detected (such as a communication base station and an electric tower) are recorded, and then in the replay stage, the movable platform automatically and regularly shoots the target to be detected according to the shooting position, the shooting height and the focal length, so that the automatic and regular monitoring of the target to be detected is realized, and the monitoring efficiency is improved.

The target tracking task refers to a tracking range of a target to be detected (such as a movable target of a person, an animal, a vehicle and the like), and a movable platform (such as an unmanned aerial vehicle) automatically tracks the target to be detected so as to monitor the real-time position of the target to be detected, observe the action, the state and the like of the target to be detected.

The target object needs to be positioned, focused and shot both in the replay stage of the target detection task and the tracking stage of the target tracking task. In some cases, because the movable platform automatically shoots under the condition of unmanned control, parameters such as a shooting position, a shooting height, a shooting focal length and the like which are input in advance are easy to deviate from actual conditions, or the problems such as shaking of a machine body caused by weather factors such as strong wind, rain and snow and the like are easy to cause, and under the condition of unmanned intervention, no target object is caused in an acquired image, or only partial images, blurred images and the like of the target object are acquired, so that unmanned automatic shooting task fails.

Taking unmanned aerial vehicle power inspection as an example: the inspection needs to check the safety condition of the key parts of the electric tower regularly, so the inspection needs to be repeated regularly, and the automatic inspection can greatly improve the efficiency and the accuracy. The automated power inspection can be divided into a teaching mode and a replay mode. The teaching mode is full-manual or semi-automatic operation, normal electric power inspection tasks are carried out, and in the process, each navigation point can record data such as the gesture of an airplane and a cradle head at each photographing (video) point. And the replay mode carries out full-automatic flight. According to the navigation point data stored in the teaching mode, the aircraft flies to each navigation point in sequence, and according to the navigation point data, the posture of the cradle head is adjusted, and corresponding photos/videos are shot.

In the existing inspection scheme, replay mainly follows a target detection flow, namely, targets are detected once, and then the positions and the focal lengths of the cradle head and the camera are directly adjusted to shoot the targets. However, in actual working, the control flow takes a long time, if a large accumulated deviation occurs in the control of the cradle head or the body is obviously swayed due to external force factors such as overlarge wind force in the process, and the long focus is sensitive to the control deviation and the body swaying, the final shooting result target may be obviously deviated from the center of the picture or even not in the picture. That is, the existing teaching/flying inspection scheme cannot solve the problem of control deviation of the cradle head and the problem of shaking of the machine body caused by strong wind and other factors. Therefore, the stability enhancement of the cradle head plays an important role in the reliability of target shooting.

In the embodiment of the disclosure, during replay, the accurate position of the inspection target in the picture is detected by using a characteristic point matching method under a low-magnification focal length to serve as an accurate photographing initialization frame, then the position of the target in the picture is tracked by using a target tracking algorithm while the posture and zooming of the tripod head are adjusted, and the tripod head control is corrected in real time to ensure that the target is always kept in the center of the picture, so that the tripod head tracking of the target is realized, and the effect of stabilizing the tripod head is achieved. And shooting the target when zooming is completed and the target is kept in the center of the picture, and finishing the accurate shooting process of the patrol target. The process of searching the target under the small focal length, carrying out pan-tilt tracking on the target and then shooting the target under the large focal length enables the aircraft to stably, accurately and clearly shoot the target details under a longer distance, and achieves a good inspection effect.

Next, each step of the image capturing method 100 will be described in detail.

In step S102, a first target image is acquired at a first preset focal length, and a target object is identified in the first target image.

In the embodiment of the disclosure, when the image capturing method 100 is applied to the replay stage of the target detection task, the second preset focal length and the position and the posture of the movable platform when the first target image is acquired are obtained according to the teaching data of the teaching stage of the target detection task. When the image capturing method 100 is applied to the tracking stage of the target tracking task, the first preset focal length may be obtained according to the tracking range of the target tracking task.

When the method of the embodiment of the disclosure is applied in the replay stage, the first preset focal length can be a focal length smaller than the second preset focal length for shooting set in the teaching stage, so that a larger shooting visual field can be obtained through the smaller focal length, and real-time positioning of the target object is facilitated. The position (including longitude, latitude and height) and the posture (including lens orientation) of the movable platform when the first target image is acquired are obtained according to teaching data of the teaching stage.

When the method of the embodiment of the disclosure is applied in the tracking stage, the first preset focal length can be obtained according to the tracking range of the target tracking task, and the position and the gesture of the movable platform during shooting can also be obtained according to the tracking range of the target tracking task. For example, in one embodiment, when the target object to be tracked is set at the point a and the photographing height is greater than x meters from the target object to prevent the target object from being found (x is a value greater than zero), the first preset focal length may be calculated according to the value of x under the condition of ensuring the photographing viewing angle as large as possible, so that the first target image is acquired under the maximum viewing angle which is centered at the point a and can be achieved by the movable platform, and the target object is repositioned by the first target image to prevent inaccurate positioning of the target object or the target object from moving during information transfer and positioning of the movable platform.

Referring to fig. 2, in one embodiment, step S102 may include:

step S1021, obtaining feature points of the target object according to the standard feature image of the target object;

step S1022, performing feature point recognition in the first target image according to the feature point of the target object, so as to determine a positioning frame of the target object.

The feature points of the target object may be acquired in advance from the standard feature image of the target object. In the replay stage of the target detection task, the feature points of the target object can be extracted according to the standard feature images of the target object provided in the teaching stage; in the tracking stage of the target tracking task, feature points of the target object can be extracted according to a standard feature image of the target object given by the target tracking task.

In the embodiment of the disclosure, a convolutional neural network may be used to extract local feature points of the current image, and the extracted local feature points are compared with feature points of the target object to identify the target object.

In the embodiment shown in fig. 2, the target object may be identified in the current image by a trained neural network model. The neural network model may be, for example, a convolutional neural network model (Convolutional Neural NetworkS, CNN). Based on the learning property of CNN, the CNN is utilized to extract characteristic points, so that the target objects with local non-texture, non-abundant textures or repeated texture characteristics such as electric tower insulators can be better processed, and the positions of the target objects can be accurately obtained. By using the CNN model to perform local feature point matching, the method can better cope with irregular-shape targets and reduce false detection caused by background matching compared with the traditional target recognition method based on image blocks.

In other embodiments of the present disclosure, the target object may also be identified using an image block detection method, or a target identification algorithm such as Scale-invariant feature transform (Scale-invariant feature transform, SIFT) algorithm, accelerated robust feature (Speeded Up Robust Features, SURF) algorithm, rapid feature point extraction and description (Oriented FAST and Rotated BRIEF, ORB) algorithm, or target detection by a variety of deep learning algorithms, which is not particularly limited in this disclosure.

In one implementation, feature point identification may be performed starting from a center point of the first target image. When the first target image is acquired, the current pose is assumed to accurately acquire the image of the target object, namely, the target object is assumed to be positioned at the center of the first target image, so that the recognition efficiency can be improved by starting from the center point of the first target image to perform feature point recognition.

In step S104, when the target object is identified in the first target image, the position and posture of the movable platform are adjusted according to the position of the target object in the first target image, so that the position of the target object is located at a preset position in the image captured by the image capturing device.

The preset position may include, for example, a central area of the shot image, and the central area may be, for example, a smaller area within a preset length and width range around a central point of the shot image, and the shape of the central area may be, for example, a rectangle or a circle.

Referring to fig. 3, in one embodiment, step S104 may include:

step S1041, a first coordinate of a target object is determined in the first target image by taking the center of the first target image as the origin of coordinates;

step S1042, determining a position and posture adjustment value of the movable platform according to the first coordinate, wherein the position and posture adjustment value comprises at least one of a horizontal adjustment value, a vertical adjustment value and a rotation angle adjustment value;

step S1043, adjusting the position and the posture of the movable platform according to the position and the posture adjustment value.

When the target object is identified in the first target image by using the local feature point comparison method, an identification frame of the target object can be obtained, and the coordinates of the identification frame in the first target image are determined as the coordinates of the target object by taking the center of the first target image as a coordinate source point, so as to determine the first coordinates of the target object.

Next, position and attitude adjustment values for the movable platform may be determined from the first coordinates. For example, when the first coordinate is (-50, 10), the movable platform may be controlled to move 50 coordinate units in the positive direction of the X-axis and 10 coordinate units in the negative direction of the Y-axis so that the target object coincides with the origin of coordinates, i.e., the center of the first target image, and the target image is located at the center position of the next frame image. The proportional relationship between the coordinate unit and the movable platform may be determined according to a scale used by the current coordinate system, which may be determined according to the flying height of the movable platform or the distance between the movable platform and the target object, which is not particularly limited in the present disclosure.

In addition to adjusting the horizontal adjustment value and the vertical adjustment value, in one embodiment, the shooting angle of the movable platform may be adjusted, that is, the rotation angle of the movable platform may be controlled, so as to shoot the target surface of the target object as much as possible, for example, the surface of the electric tower on which the key facilities are disposed, the surface of the tracking target, and the like. The rotation angle adjustment value of the movable platform can be judged according to the characteristic recognition result of the target object so as to determine the angle difference between the side surface of the target object photographed at present and the standard side surface to be photographed, and the rotation angle adjustment value of the movable platform is obtained through conversion according to the scale of the current coordinate system and the angle difference.

In another embodiment, a distance adjustment value of the movable platform from the target object may also be determined. For example, when the unmanned aerial vehicle is performing nodding on the target object, the flying height of the fuselage becomes high due to the influence of strong wind, and at this time, the size of the recognition frame of the target object in the current image becomes smaller than that of the teaching stage or the preset recognition frame, so that the flying height of the movable platform can be readjusted according to the set flying height, or a distance adjustment value is obtained by conversion according to the proportional relation between the size of the recognition frame and the preset recognition frame standard value, and the unmanned aerial vehicle is controlled to approach or depart from the target object. The variety of the position and posture adjustment values of the movable platform can be various, and the movable platform can be set by a person skilled in the art according to actual conditions.

When the camera device is directly erected on the movable platform, the movable platform can be controlled to adjust the position and the posture so as to lead the shooting center of the camera device to be aligned to the target object. When the camera device is supposed to be on the cradle head of the movable platform, the shooting center of the camera device can be aligned to the target object by adjusting the position and the posture of the cradle head.

After pose adjustment is performed on the movable platform, the current shooting center is defaulted to be aligned to the target object, and the focal length can be adjusted to be a second preset focal length so as to shoot the target object.

The second preset focal length is a preset focal length value capable of accurately observing the target object. In the replay stage, the second preset focal length may be, for example, an ideal shooting focal length set in the teaching stage; in the tracking stage, the second preset focal length is, for example, a preset tracking shooting focal length value.

In the embodiment of the disclosure, in the process of adjusting the position and the posture of the movable platform according to the position of the target object in the first target image, the target object is continuously tracked, so that the target object is continuously in the picture shot by the camera.

In one embodiment, the method may also be set when the target object is located in the central area of the current image, and the focal length is directly adjusted to a second preset focal length and photographed without pose adjustment, so as to reduce the calculation amount and improve the calculation efficiency.

In step S106, when the target object is not recognized in the first target image, the focal length of the image capturing device is continuously increased, and the target object is recognized according to the current image collected by the image capturing device until the focal length of the image capturing device is equal to the second preset focal length. And in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so that the position of the target object is positioned at a preset position in a picture shot by the image pickup device.

In an embodiment of the present disclosure, continuously increasing the focal length of the image capturing apparatus includes: continuously increasing the focal length of the image pickup device by a preset step length.

The current image acquired by the image capturing apparatus in step S106 is not a captured image, but buffered data of a real-time field of view of the image capturing apparatus. The current image is used to aid in image recognition and location analysis, and is deleted in a short time. Therefore, when the embodiment of the disclosure is executed by the processor, the processor acquires the current image in real time and analyzes the current image, and identifies the target object, so as to adjust the pose of the movable platform while continuously zooming according to the position of the target object in the current image, so that the target object is located at a preset position in the image captured by the image capturing device, where the preset position is, for example, a central area. The central area may be, for example, a smaller area within a preset length and width range around the central point of the current image, and the shape of the central area may be, for example, a rectangle or a circle.

Referring to fig. 4, in one embodiment, step S106 may include:

step S1061, obtaining feature points of the target object according to the standard feature image of the target object;

Step S1061, performing feature point identification on the current image acquired by the image capturing device according to the feature point of the target object, so as to determine a positioning frame of the target object.

The process of identifying the target object in step S106 is similar to the process of identifying the target object in step S104, and the feature point identification can be performed based on various algorithms such as Convolutional Neural Network (CNN), and will not be described herein.

When the feature is identified, feature point identification can be performed from the center point of the current image, so that the identification efficiency is improved. Feature point recognition from the center point of the current image may be represented as recognition near the recognition frame position of the previous frame image to determine the recognition frame of the target object in the current image.

In one embodiment, in step S1061, feature points of the target object may also be obtained according to image features in the recognition frame of the target object in the previous frame of image, so as to improve the accuracy of information of the target object based on the latest information and improve the recognition accuracy.

The robust automatic inspection scheme based on feature matching positioning targets and cradle head tracking can solve the problems of automatic accurate positioning and stable shooting of targets such as electric towers. In order to solve the problem of shooting deviation caused by control errors of a tripod head or strong wind in the time of tripod head position adjustment and zooming, firstly, a precise shooting target matching frame is used as an initialization frame to track the target, so that the deviation in the tripod head control and zooming processes is overcome, the tripod head stability increasing effect is achieved, the target in the control process is always kept in the center of a picture until the zooming is finished, and a complete and clear inspection target is shot. Compared with the traditional target detection method, the CNN-based local feature extraction method is used for extracting the local feature and can be better suitable for the situation of local non-texture or repeated textures.

According to the method provided by the embodiment of the disclosure, the initialization frame for accurately shooting is obtained based on the local feature matching positioning target, and the cradle head stability enhancement in the replay process is realized by combining the cradle head tracking method of the target, so that the accurate and robust inspection target shooting effect is finally achieved, and the method can be used for solving the problem that the target is not in the center of a picture or in the visual field due to insufficient positioning of a camera (airplane), insufficient control of the cradle head or shaking of a machine body due to external factors such as strong wind of the camera (airplane) in the automatic inspection process, and can be applied to an industrial unmanned aerial vehicle with the inspection function.

The cradle head stability augmentation inspection scheme combining the accurate photographing initialization frame and cradle head tracking is an automatic inspection scheme with closed-loop control. When the inspection target is replayed and shot, a target frame in a short-focus picture is firstly obtained by using a characteristic point matching method to serve as an initialization frame for accurately shooting the inspection target, and then the target is subjected to pan-tilt tracking by using a target tracking method so as to resist deviation caused by accumulated control errors or strong wind and other external forces possibly occurring in the pan-tilt pose adjustment and zooming processes.

Referring to fig. 5, an image photographing method 500 may include:

step S501, shooting a replay image in a short focus;

step S502, identifying a target object through a CNN algorithm;

step S503, extracting feature points and descriptors of the target object;

meanwhile, a target teaching graph is acquired in step S501';

identifying a target object through a CNN algorithm at step S502';

extracting feature points and descriptors of the target object at step S503';

next to this, the process is carried out,

in step S504, feature points are matched;

step S505, calculating a short-focus repeated shooting picture target frame;

step S506, adjusting the pose and zooming of the cradle head;

step S507, tracking;

step S508, judging whether the zooming and the control are finished, if yes, entering step S509, shooting the inspection target by using the long focus, and then finishing the flow, and if yes, returning to step S506, and readjusting the pose of the cradle head and the zooming.

In step S501, the short-focus shooting of the replay image refers to capturing an image at a first preset focal length, which belongs to real-time observation of a target object. In step S501', a teaching image of the target object, i.e., a target teaching chart, is obtained from the teaching data.

In steps S502, S503 and steps S502', S503', the target object is identified in the two pictures (the replay image and the teaching image) respectively using the CNN algorithm, and the feature points and descriptors of the target object in the two pictures are extracted.

The characteristic points extracted by using the CNN can better cope with targets with not abundant textures or repeated texture characteristics such as electric tower insulators and the like based on the learning property of the CNN. And compared with the method based on the image block, the target positioning method based on the local feature point matching can better cope with the irregular shape target, and the problem of false detection caused by background matching is reduced. The feature point matching may use a local feature extraction method, or may use a conventional method, including but not limited to SIFT, SURF, ORB algorithm, etc.

In step S504, feature point matching is performed on the feature points and descriptors extracted from the two pictures to calculate a target frame in the short Jiao Fu shot image, also referred to as an extraction initialization frame, in step S505 according to the matching result. The initialization frame acquisition method may be replaced with an image block-based detection method, not limited to the conventional method or the deep learning method.

Step S506 to step S508 are processes of adjusting the pose of the pan-tilt by loop iteration.

In step S506, zooming refers to continuously increasing the focal length, for example, the focal length may be increased according to a preset compensation; the adjustment of the pan-tilt pose is, for example, adjustment of the pan-tilt pose on parameters such as horizontal direction, vertical direction, deflection angle, etc., so as to adjust the shooting angle. The adjustment of the cradle head pose and zooming can be performed simultaneously, and the camera device continuously collects real-time images.

In step S506, target tracking is performed in step S507 based on the image continuously acquired by the imaging device and the position recognition result of the target object in the image. The target tracking can be completed directly by the feature point matching method for acquiring the initialization frame, namely, the position of the target frame is updated by the feature point matching method for each frame (or a plurality of frames at intervals). During cradle head tracking (track), features can be extracted in a target frame of a previous frame, then feature searching and matching are carried out near the old frame position of the current frame, the position of the target frame is updated, the cradle head is controlled to enable a drawing target to be kept in the center of a picture, the cradle head stability increasing effect is achieved, and the process is iterated until zooming and control are finished. The basis for the end of zooming and controlling is that the preset focal length is reached and the target frame is still in the center of the picture, and at the moment, the shooting of the target is completed, and the whole inspection target with the cradle head effect is accurately shot again.

After judging in step S508 that the zooming and controlling are finished, the flow proceeds to step S509 to photograph the patrol target using the tele (second preset focal length) and then to finish the flow.

According to the embodiment of the disclosure, the CNN-based characteristic point detection method is utilized to obtain the accurate photographing initialization frame, the cradle head tracking method of the target is utilized to perform cradle head stability augmentation, the closed-loop control of inspection target photographing is realized, photographing deviation caused by cradle head control errors, strong wind and other external factors in the zooming and cradle head pose adjusting process is overcome, and the method has the advantages that:

1. Based on the learning property of the CNN, the CNN can extract local features to better adapt to the situation of local non-texture or repeated textures, and accurately obtain the target frame position;

2. compared with a shooting mode of directly controlling the pose and zooming of the cradle head by one-time target detection, the target tracking of the cradle head is utilized to keep the target at a position in the center of a picture all the time in the whole control process, so that the target shooting deviation caused by external forces such as cradle head control or strong wind can be better dealt with, and even the target is lost.

Referring to fig. 6, in one embodiment, the entire process of the image photographing method may include:

in step S601, a first image is acquired at a first preset focal length, the first image is configured as a current image, and the first preset focal length is configured as a current focal length.

Step S602, a target object is identified in the current image.

Step S603, it is determined whether or not the target object is identified, and if the target object is identified, the process proceeds to step S604, otherwise the process proceeds to step S613.

Step S604, the current feature point of the target object is updated according to the identification result of the target object in the current image.

Step S605, judging whether the target object is positioned at the center of the current image, if so, proceeding to step S606, adjusting the pose according to the coordinate difference between the target object and the center of the current image, and proceeding to step S607; if not, the process proceeds directly to step S607.

Step S607, determining whether the current focal length is equal to the second preset focal length, if so, proceeding to step S608; otherwise, the process advances to step S608, where the current focal length is increased, a second image is acquired, and the second image is configured as the current image, and the process returns to step S602.

In step S609, a third image is acquired, and the third image is configured as the current image.

Step S610, judging whether the target object is positioned at the center of the current image, if not, proceeding to step S611, after adjusting the pose according to the coordinate difference value between the target object and the center of the current image, returning to step S609 until the target object is positioned at the center of the current image; if so, the process advances to step S612 to photograph the target object.

Step S613, if the target object is not identified in the current image, judging whether the current focal length is equal to the third preset focal length, if not, proceeding to step S614, reducing the current focal length, acquiring a fourth image, configuring the fourth image as the current image, and returning to step S602; if so, the process advances to step S615 to output identification failure information.

In the embodiment shown in fig. 6, steps S601 to S608 are iterative zooming methods when the target object can be directly recognized in the first image. In step S601, the current image is a parameter, not a unique image, which can be assigned values to equal different images at different times; the current focal length is likewise a parameter, not a unique value, which may also be assigned to equal different focal length values at different times.

The method for identifying the target object in step S602 and step S603 is described in the above embodiments, and is not described herein.

In step S604, the current feature point of the target object is a parameter, when the current image is equal to the first image, the current feature point of the target object is equal to the feature point of the target object in the teaching image, and when the current image is equal to the other image, the current feature point of the target object that has not been updated is equal to the feature point of the target object identified in the previous frame image. Therefore, in step S604, the parameter of the feature point of the target object is updated in real time according to the feature point of the identified target object, so that the feature point of the target object can be kept obtained based on the latest identification data, and the recognition error caused by the discrepancy between the teaching data and the real-time situation can be reduced.

In steps S605, S606, if the target image is not located at the center of the current image (or the center area as described in the foregoing embodiment), the position and posture of the pan-tilt or unmanned aerial vehicle may be adjusted according to the coordinate difference of the target object and the center of the current image. The coordinates of the target object may be calculated with the center of the current image as the origin.

In step S607, if the target object is located at the center of the current image, or after pose adjustment, the target image is located at the center of the current image, it may be determined whether the current focal length reaches the set photographing focal length (the second preset focal length), if not, step S608 is entered to continue increasing the focal length to enlarge the duty ratio of the target object in the image, improve the photographing definition of the target object, collect the second image at the newly increased focal length, and perform pose adjustment according to the photographed image at the larger focal length in the manner of step S602 to step S607, so that the target object can be located at the photographing center of the current image at the larger focal length until reaching the preset photographing focal length, that is, the second preset focal length.

Step S609 to step S612 are shooting fine adjustment under a second preset focal length, so as to complete the shooting process.

If fine adjustment of photographing is not required, a third image may be directly acquired at the second preset focal length in step S609, and the third image may be saved as a photographing result of the target object.

However, in some cases, when the current focal length is equal to the second preset focal length after the focal length is increased, the second image captured at the second preset focal length is set as the current image, and thus, after the pose (of the unmanned aerial vehicle or the cradle head) is adjusted according to the target object in the current image in step S606, the process proceeds directly to step S609 through the judgment in step S607 to capture the third image, and the result after the pose adjustment is not checked yet, that is, whether the pose adjustment is in place or not.

Therefore, in step S610, it may be determined that the third image acquired (not necessarily photographed) after the pose adjustment is performed at the second preset focal length, and if the pose adjustment is not in place (the target object is not located at the center of the current image/the third image), step S611 is performed to continue adjusting the pose until the pose adjustment is in place, and step S612 is performed to photograph.

Before proceeding to step S612 to capture the target object, the acquired first image, second image, and third image may be image data that changes in real time in the lens when the camera has not pressed the capture key.

Steps S613 to S615 are one processing method in the case where the target loss is found. In some cases, such as when a strong wind causes a body to shake, there may be no target object in the acquired target images, whether the first image is acquired based on the first preset focal length or the second image is acquired based on the increased focal length. At this time, the current focal length may be reduced to increase the shooting field of view, and the positioning target object may be re-recognized. When the current focal length is reduced to the third preset focal length, if the target object cannot be identified in the current image, identification failure information can be output, and the target loss is reported. The third preset focal length is smaller than the first preset focal length, and the third preset focal length can be set by a person skilled in the art according to the shooting capability of the shooting device.

The method provided in the embodiment shown in fig. 6 firstly controls the movable platform to shoot according to the set position and height under a smaller first preset focal length, so as to ensure that the target object is in the lens first, then adopts the iterative zoom positioning method, does not determine the target object, and finally shoots the target object under a second preset focal length capable of ensuring shooting definition.

Through the loop iteration of the embodiment shown in fig. 6, the characteristic information and the position information of the target object can be updated according to the image acquired in real time, so that the ideal shooting pose and the shooting focal length can be gradually approached, and the shooting failure of the target object caused by the deviation of the control parameters input in advance or the shaking problem of the body caused by strong wind and other external forces can be avoided.

The cradle head stability augmentation inspection scheme combining the accurate photographing initialization frame and cradle head tracking is an automatic inspection scheme controlled in a closed loop mode. When the inspection target is replayed and shot, a target frame in a short-focus picture is firstly obtained by using a characteristic point matching method to serve as an initialization frame for accurately shooting the inspection target, and then the target is subjected to pan-tilt tracking by using a target tracking method so as to resist deviation caused by accumulated control errors or strong wind and other external forces possibly occurring in the pan-tilt pose adjustment and zooming processes. When the cradle head is used for tracking (track), the current characteristic point of the target object is firstly extracted from the previous frame identification frame, then characteristic searching and matching are carried out near the old frame position of the current frame, the identification frame position is updated, the cradle head is controlled to enable the target object to be kept in the center of a shooting picture, the cradle head stability increasing effect is achieved, and the process is iterated until zooming and control are finished. The basis for the end of zooming and controlling is that the preset focal length is reached and the target frame is still in the center of the picture, and at the moment, the shooting target finishes the accurate re-shooting of the inspection target. Compared with a shooting mode of directly controlling the pose and zooming of the tripod head by one-time target detection, the method has the advantages that the target is always kept at a relatively central position of a picture in the whole control process by utilizing the target tracking of the tripod head, target shooting deviation caused by external force such as tripod head control or strong wind and even target loss can be better dealt with, tripod head stability enhancement in the replay process is realized, and finally, the accurate and robust inspection target shooting effect is achieved.

Referring to fig. 7, a movable platform 700 may include:

a body 71;

a power system 72, provided to the machine body, for powering the movable platform;

the camera 73 is arranged on the machine body and is used for collecting images;

a memory 74; and

a processor 75 coupled to the memory, the processor being configured to execute the image capturing method of the embodiment shown in fig. 1 to 6 based on instructions stored in the memory.

In one exemplary embodiment of the present disclosure, the processor 75 controls the power system 72 to adjust the position and posture of the movable platform 700 when adjusting the position and posture of the movable platform 700 in the course of performing the image photographing method.

In another embodiment of the present disclosure, the movable platform 700 further includes a cradle head 76, and the image capturing device 73 is mounted on the cradle head 76. In adjusting the position and posture of the movable platform 700 during the execution of the image photographing method, the processor 75 controls the subsystem 72 to adjust the position and posture of the pan-tilt 76 to keep the target object at the photographing center.

The method and the device can be used for an industrial unmanned aerial vehicle with a patrol function, and solve the problem that a target is not in the center of a picture or in the field of view due to insufficient positioning of a camera (airplane) and insufficient control of a cradle head in the automatic patrol process, and the problem that the target is not in the center of the picture or in the field of view due to shaking of a machine body caused by external force factors such as strong wind of the camera (airplane) in the automatic patrol process.

Corresponding to the above method embodiments, the present disclosure also provides an image capturing apparatus, which may be used to perform the above method embodiments.

Referring to fig. 8, an image photographing device 800 may include:

a memory 81 configured to store program codes;

one or more processors 82 coupled to the memory 81, the processor 82 being configured to perform the following methods based on instructions stored in the memory 81:

collecting a first target image at a first preset focal length, and identifying a target object in the first target image;

when the target object is identified in the first target image, adjusting the position and the posture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the shooting device;

when the first target image cannot identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length;

when the target object is identified according to the current image acquired by the image pickup device at any moment in the process of adjusting the focal length of the image pickup device, the position and the gesture of the movable platform are adjusted according to the position of the target object in the current image, so that the position of the target object is located at a preset position in a picture shot by the image pickup device.

In an exemplary embodiment of the present disclosure, the second preset focal length and the position and the posture of the movable platform when the first target image is acquired are obtained according to teaching data of a teaching stage of the target detection task.

In one exemplary embodiment of the present disclosure, the first preset focal length is obtained according to a tracking range of the target tracking task.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: determining a first coordinate of a target object in a first target image by taking the position of a central area of the first target image as a coordinate origin; determining a position and posture adjustment value of the movable platform according to the first coordinate, wherein the position and posture adjustment value comprises at least one of a horizontal adjustment value, a vertical adjustment value and a rotation angle adjustment value; and adjusting the position and the posture of the movable platform according to the position and posture adjustment values.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: acquiring feature points of the target object according to the standard feature image of the target object; and carrying out feature point identification in the first target image according to the feature points of the target object so as to determine the positioning frame of the target object.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: acquiring feature points of the target object according to the standard feature image of the target object; and carrying out feature point identification in the current image acquired by the image pickup device according to the feature points of the target object so as to determine a positioning frame of the target object.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: feature point recognition is performed starting from the center region of the first target image.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: extracting local feature points of the first target image by using a convolutional neural network; and comparing the extracted local feature points with the feature points of the target object to identify the target object.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: the focal length of the image pickup device is continuously increased by a preset step length.

In one exemplary embodiment of the present disclosure, the processor 82 is configured to: and continuously tracking the target object in the process of adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so as to ensure that the target object is continuously in the picture shot by the camera.

It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.

Those skilled in the art will appreciate that the various aspects of the invention may be implemented as a system, method, or program product. Accordingly, aspects of the invention may be embodied in the following forms, namely: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, a terminal device, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.

In an exemplary embodiment of the present disclosure, a computer-readable storage medium having stored thereon a program product capable of implementing the method described above in the present specification is also provided. In some possible embodiments, the various aspects of the invention may also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps according to the various exemplary embodiments of the invention as described in the "exemplary methods" section of this specification, when said program product is run on the terminal device.

The program product for implementing the above-described method according to an embodiment of the present invention may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may be run on a terminal device such as a personal computer. However, the program product of the present invention is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The computer readable signal medium may include a data signal propagated in baseband or as part of a carrier wave with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

Furthermore, the above-described drawings are only schematic illustrations of processes included in the method according to the exemplary embodiment of the present invention, and are not intended to be limiting. It will be readily appreciated that the processes shown in the above figures do not indicate or limit the temporal order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, for example, among a plurality of modules.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Industrial applicability

Claims

An image shooting method is characterized by being applied to a movable platform, wherein an image pickup device is carried on the movable platform, and the movable platform collects images through the image pickup device, and the method comprises the following steps:

collecting a first target image at a first preset focal length, and identifying a target object in the first target image;

when the target object is identified in the first target image, adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the camera;

when the first target image does not identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length;

and in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so that the position of the target object is positioned at a preset position in a picture shot by the image pickup device.
The image capturing method according to claim 1, wherein the method is applied to a replay stage of a target detection task, and the second preset focal length and the position and the posture of the movable platform when the first target image is acquired are obtained according to teaching data of a teaching stage of the target detection task.
The image capturing method according to claim 1, wherein the method is applied to a tracking stage of a target tracking task, and the first preset focal length is obtained according to a tracking range of the target tracking task.
The image capturing method of claim 1, wherein adjusting the position and posture of the movable platform according to the position of the target object in the first target image comprises:

determining a first coordinate of the target object in the first target image by taking the center of the first target image as a coordinate origin;

determining a position and posture adjustment value of the movable platform according to the first coordinate, wherein the position and posture adjustment value comprises at least one of a horizontal adjustment value, a vertical adjustment value and a rotation angle adjustment value;

and adjusting the position and the posture of the movable platform according to the position and posture adjustment value.
The image capturing method of claim 1, wherein identifying a target object in the first target image comprises:

acquiring feature points of the target object according to the standard feature image of the target object;

and carrying out feature point identification in the first target image according to the feature points of the target object so as to determine a positioning frame of the target object.
The image capturing method according to claim 1, wherein identifying the target object from the current image acquired by the image capturing device includes:

acquiring feature points of the target object according to the standard feature image of the target object;

and carrying out feature point identification in the current image acquired by the image pickup device according to the feature points of the target object so as to determine a positioning frame of the target object.
The image capturing method according to claim 5, wherein the performing feature point identification in the first target image from the current feature point includes:

and performing feature point identification from the central point of the first target image.
The image capturing method according to claim 5 or 7, wherein the performing feature point identification in the first target image from the current feature point includes:

Extracting local feature points of the first target image by using a convolutional neural network;

and comparing the extracted local characteristic points with the characteristic points of the target object to identify the target object.
The image capturing method according to claim 1, wherein the continuously increasing the focal length of the image capturing apparatus includes: continuously increasing the focal length of the image pickup device by a preset step length.
The image capturing method according to claim 1, wherein the target object is continuously tracked in adjusting the position and posture of the movable platform in accordance with the position of the target object in the first target image so that the target object is continuously in the picture captured by the image capturing device.
The image capturing method of claim 1, wherein the movable platform comprises an unmanned aerial vehicle.
The image capturing method of claim 11, wherein the movable platform further comprises a cradle head mounted on the unmanned aerial vehicle, the camera device being mounted on the cradle head.
An image capturing apparatus, characterized by comprising:

A memory configured to store program code;

one or more processors coupled to the memory, the processors configured to perform the following methods based on instructions stored in the memory:

collecting a first target image at a first preset focal length, and identifying a target object in the first target image;

when the target object is identified in the first target image, adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so that the position of the target object is located at a preset position in a shooting picture of the camera;

when the first target image does not identify the target object, continuously increasing the focal length of the image pickup device, and identifying the target object according to the current image acquired by the image pickup device until the focal length of the image pickup device is equal to a second preset focal length;

and in the process of adjusting the focal length of the image pickup device, when the target object is identified according to the current image acquired by the image pickup device at any moment, adjusting the position and the posture of the movable platform according to the position of the target object in the current image so that the position of the target object is positioned at a preset position in a picture shot by the image pickup device.
The image capturing apparatus according to claim 13, wherein the second preset focal length and the position and posture of the movable platform at the time of capturing the first target image are obtained from teaching data of a teaching stage of the target detection task.
The image capturing apparatus according to claim 13, wherein the first preset focal length is obtained from a tracking range of the target tracking task.
The image capturing device of claim 13, wherein the processor is configured to:

determining a first coordinate of the target object in the first target image by taking the central area position of the first target image as a coordinate origin;

determining a position and posture adjustment value of the movable platform according to the first coordinate, wherein the position and posture adjustment value comprises at least one of a horizontal adjustment value, a vertical adjustment value and a rotation angle adjustment value;

and adjusting the position and the posture of the movable platform according to the position and posture adjustment value.
The image capturing device of claim 13, wherein the processor is configured to:

acquiring feature points of the target object according to the standard feature image of the target object;

And carrying out feature point identification in the first target image according to the feature points of the target object so as to determine a positioning frame of the target object.
The image capturing device of claim 13, wherein the processor is configured to:

acquiring feature points of the target object according to the standard feature image of the target object;

and carrying out feature point identification in the current image acquired by the image pickup device according to the feature points of the target object so as to determine a positioning frame of the target object.
The image capturing device of claim 17, wherein the processor is configured to:

feature point identification is performed starting from a central region of the first target image.
The image capturing apparatus according to claim 17 or 19, wherein the processor is configured to:

extracting local feature points of the first target image by using a convolutional neural network;

and comparing the extracted local characteristic points with the characteristic points of the target object to identify the target object.
The image capturing device of claim 13, wherein the processor is configured to: continuously increasing the focal length of the image pickup device by a preset step length.
The image capturing device of claim 13, wherein the processor is configured to:

and continuously tracking the target object in the process of adjusting the position and the gesture of the movable platform according to the position of the target object in the first target image so as to ensure that the target object is continuously in a picture shot by the camera.
A movable platform, comprising:

a body;

the power system is arranged on the machine body and is used for providing power for the movable platform;

the camera device is arranged on the machine body and used for collecting images;

a memory; and

a processor coupled to the memory, the processor configured to perform the image capture method of any of claims 1-12 based on instructions stored in the memory.
The mobile platform of claim 23, wherein the processor controls the power system to adjust the position and attitude of the mobile platform when adjusting the position and attitude of the mobile platform during execution of the image capture method.
The mobile platform of claim 23, further comprising a pan-tilt, the camera device being mounted on the pan-tilt; and when the position and the posture of the movable platform are adjusted in the process of executing the image shooting method, the processor controls the power system to adjust the position and the posture of the cradle head.
A computer-readable storage medium having stored thereon a program which, when executed by a processor, implements the image capturing method according to any one of claims 1 to 12.