WO2023035177A1

WO2023035177A1 - Target confirmation method and apparatus, and handheld system

Info

Publication number: WO2023035177A1
Application number: PCT/CN2021/117414
Authority: WO
Inventors: 程正喜; 胡晓翔; 封旭阳
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2021-09-09
Filing date: 2021-09-09
Publication date: 2023-03-16

Abstract

A target confirmation method and apparatus, and a handheld system. The method comprises: obtaining a recommendation target, wherein the recommendation target is determined according to feature information of a first target in a scene image; marking the recommendation target to obtain a marked target; obtaining an external instruction, and performing target confirmation on the basis of the external instruction and the marked target, wherein the external instruction is used for selecting a second target in the scene image; and performing target tracking by using the confirmed target, so that the confirmed target is kept within a predetermined range of a screen, and the screen is configured to display the scene image and the marked target. The method can solve the problem in the prior art that when a user selects a target of interest, the user can only manually select the target of interest on the screen, thereby causing being not intelligent enough and affecting the user experience.

Description

A target confirmation method, device and handheld system

technical field

The present application relates to the technical field of video shooting, and in particular to a target confirmation method, device and handheld system.

Background technique

With the leapfrog development of shooting technology, people have a higher demand for continuous tracking and shooting of specific targets. From the original human tracking to the intelligent tracking of electronic equipment, the target tracking technology is becoming more and more perfect.

Usually, when a user uses an electronic device to track and shoot a specific target, he often needs to select a specific target of interest on a display screen in advance.

Contents of the invention

The embodiment of the present application provides a target confirmation method, device and handheld system, which are used to at least solve the lack of intelligence caused by manually selecting the target of interest on the screen when the user selects the target of interest in the prior art , issues affecting user experience.

In the first aspect, the embodiment of the present application provides a target confirmation method, including:

Acquiring a recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Marking the selected target to obtain the marked target;

Acquiring an external instruction, and performing target confirmation based on the external instruction and the marked target, the external instruction is used to select a second target in the scene image;

The confirmed target is used for target tracking, so that the confirmed target remains within a predetermined range of a screen for displaying the scene image and the marked target.

In a second aspect, an embodiment of the present application provides a target confirmation device, including: a memory, a processor, and a screen;

said memory for storing a computer program capable of running on said processor;

The processor is configured to perform the following operations when running the computer program:

Marking the selected target to obtain the marked target;

performing target tracking by using the confirmed target, so that the confirmed target remains within a predetermined range of the screen;

The screen is used to display the scene image and the marked target.

In the third aspect, the embodiment of the present application provides a handheld system, including: a handheld device, a memory, a processor, and a screen, the memory and the processor are arranged in the handheld device, and the screen communicates with the handheld device connect;

The hand-held device includes a hand-held part for being held by a user;

Marking the selected target to obtain the marked target;

The screen is used to display the scene image and the marked target.

It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

Embodiments of the present application provide a target confirmation method, device, and handheld system. After obtaining the selected target according to the feature information of the first target in the scene image, the obtained selected target can be marked to associate the selected target with the scene. The other targets in the image are distinguished, and the user can select the second target in the scene image by issuing an external command. Finally, the final tracked target can be confirmed based on the marked target and the external command, and the tracked target is always Stay within the bounds of the screen presets. This target tracking method is different from the existing technology that can only manually select the target to be tracked by the user touching the screen, but it can intelligently recommend and mark the target that the user may be interested in, and then combined with external instructions to confirm the final use for tracking The goal. Therefore, when the user needs to confirm the target, the machine can intelligently recommend some targets for the user, which improves the user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort.

Fig. 1 is the basic frame diagram of two-handed pan-tilt;

Fig. 2 is a schematic flow chart of a target confirmation method provided by an embodiment of the present application;

FIG. 3 is a schematic structural diagram of a target confirmation device provided by an embodiment of the present application;

Fig. 4 is a schematic structural diagram of a target confirmation device provided in another embodiment of the present application.

Explanation of reference signs:

11. No. 1 hand-held part; 12. No. 2 hand-held part; 13. Main body of the gimbal; 14. Screen;

40. Handheld system; 41. Handheld device; 42. Memory; 43. Processor; 44. Screen.

Detailed ways

In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiment of the application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiment of the application. Obviously, the described embodiment is only It is an embodiment of a part of the application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present application and the above drawings are used to describe similar objects and do not necessarily use Used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

The development of shooting technology has made people have higher requirements for shooting. In the field of image or video shooting, there are many situations that need to confirm the target of interest, such as tracking and shooting the target, and interested in the captured image. Focus on the target or zoom in on the target of interest. However, in the current prior art, the target confirmation is usually performed by the user touching the screen of the device, and this target confirmation method is not smart enough. In order to make the target confirmation method more intelligent and improve user experience, the embodiments of the present application provide a target confirmation method, device and handheld system.

Exemplarily, in this embodiment of the application, some targets can be selected and marked for the user according to the feature information of the target in the scene image, and the user can issue an external command to select the target in the scene, and then combine the marked target with the external The command confirms the target to be tracked, and the final confirmed target will always be kept within the predetermined range of the screen to complete the target tracking task. Different from the prior art where the user confirms the target of interest by touching the screen, the method provided in the embodiment of the present application can intelligently recommend the target that the user may be interested in, combined with external instructions, the recommended target can be Confirm the final target of interest, you can also confirm the final target of interest in addition to the recommended target, enrich the confirmation method of tracking target, and solve the problem that the user cannot select the target of interest by touching the screen in some special cases The problem.

Exemplarily, a method for confirming a target provided in the embodiment of the present application may include but not limited to be applied to a scene where a mobile platform or a mobile platform is used in cooperation with a control device. Among them, the mobile platform includes devices that can move freely or artificially and collect external images, such as handheld pan-tilts, unmanned aerial vehicles, unmanned vehicles, and unmanned ships. The control device can be used to control the movable platform, and the control device can be provided with a screen, which can display some targets in the acquired scene image and the recommended scene image. The control device includes but is not limited to a remote When using a controller, the screen can be integrated into the remote control, or it can be designed separately from the remote control), mobile phone, tablet computer, wearable device, etc. Among them, when applied to a mobile platform, the mobile platform can be equipped with a screen, and the screen can be integrated in the mobile platform or separately installed on the mobile platform, and the user can issue an external command based on the display content of the screen, so as to follow the recommended target Confirm the target that needs to be tracked and shot with the actual target that you are interested in; when it is applied to the scene where the movable platform is used in conjunction with the control device, the movable platform is far away from the control device, and the user can send a message based on the display content of the screen at the control device. The external instruction confirms the final target that needs to be tracked and shot based on the recommended target and the actual target of interest.

Exemplarily, a typical application scenario of a target confirmation method and device provided in the embodiments of the present application is a handheld system, especially a dual-handheld system, such as a dual-handheld pan-tilt, and a dual-handheld remote controller used in conjunction with a mobile platform. In particular, when the method provided by the embodiment of the present application is applied to a two-handed system, it can bring additional beneficial effects. Here, the application of the method and device provided by the application to the two-handed pan-tilt is described as an example. As shown in FIG. No. 2 hand-held part 12 , wherein No. 1 hand-held part 11 and No. 2 hand-held part 12 are connected to both sides of the pan-tilt main body 13 , and the screen 14 is detachably connected to the pan-tilt main body 13 . When the user uses the two-handed pan/tilt as shown in Figure 1, the user's hands hold No. 1 hand-held part 11 and No. 2 hand-held part 12 respectively, and the image acquisition device carried by the pan/tilt main body 13 shoots the object to be photographed, and The captured image is transmitted to the screen 14 to present the captured scene image, and the user can monitor the captured scene through the screen 14 and also adjust the shooting parameters. In practical application, the user can use the two-handed pan-tilt by holding the No. 1 hand-held part 11 and No. 2 hand-held part 12 as shown in Figure 1, but when confirming the tracking target, the user needs to free up one hand to On the screen 14 where the scene image is displayed, the target of interest is confirmed by means such as touch, frame selection, and the like. However, under normal circumstances, holding the pan-tilt with two hands in one hand often brings an inconvenient experience to the user. For example, when the user is holding the two-handed pan-tilt with both hands, a target that needs to be tracked and shot suddenly appears in the scene image captured by the image acquisition device. If the user only holds the pan-tilt with one hand and frees up One hand touches and selects the target to be tracked on the screen 14, and the pan-tilt itself often has a certain weight due to the need for both hands. It brings a certain amount of inconvenience, such as the inability to use the two-handed gimbal stably, resulting in shaking of the two-handed gimbal and making the scene image abnormal. However, when the method provided by the embodiment of the present application is applied to the two-handed pan-tilt as shown in Figure 1, in the scene where the user needs to confirm the target, the scene image and the scene image selected for the user can be displayed on the screen 14. objects of possible interest. The user can send an external command to confirm the target of his own interest among the selected targets without freeing up a hand to touch the screen 14, for example, the user can click on the No. The area that can be operated by the finger selects the target of interest by controlling the operating part (such as selecting and switching the target of interest in the selected target by pressing a button or a knob), and the user can also select the target of interest through his own gesture, or Users can also select objects of interest by issuing voice commands. In this way, when the method provided by the embodiment of the present application is applied to the two-handed pan/tilt, it can not only bring a novel target confirmation mode to the user, but also does not require the user to free up one hand to select the target, which brings certain benefits. Convenience and improved user experience.

It should be noted that the beneficial effects brought by the method provided by the present application are only described above by taking the two-handed pan-tilt as an example, but it does not mean that only the application of the method provided by the application on the two-handed pan-tilt can bring For the above beneficial effects, it should be understood that when the method provided by the present application is applied to a dual handheld device similar to a dual handheld pan/tilt, it will also bring the same or similar beneficial effects.

For ease of understanding, the following describes the specific processes in the embodiments of the present application. If there is no conflict, the features in the following embodiments may be coupled or decoupled to form different technical solutions.

Executors of all the method embodiments provided in this application include but are not limited to a mobile platform or a control device used in conjunction with the mobile platform, specifically a processor that may be provided therefor. Fig. 2 is a schematic flowchart of a target confirmation method provided by an embodiment of the present application. As shown in Figure 2, the method of this embodiment can include:

Step S0201, acquiring a recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Step S0202, marking the selected target to obtain the marked target;

Step S0203, acquiring an external command, and performing target confirmation based on the external command and the marked target, the external command is used to select a second target in the scene image;

Step S0204, using the confirmed target to track the target, so that the confirmed target remains within a predetermined range of the screen, and the screen is used to display the scene image and the marked target.

The method provided in this embodiment can be specifically described as: Obtaining the first target in the scene image according to the target recognition algorithm, for example, after using the target contour extraction algorithm to extract the target area in the scene image, and then using the deep learning algorithm to identify the target area in the target area target identification. In order to further more accurately select targets that may be of interest to the user, the processor selects some targets from the first targets for the user according to at least one feature information of the first target. In order to distinguish the selected target from other targets in the scene image, the processor can use the identifier to mark the selected target to obtain the marked target, and display the marked target and the scene image on the screen. In addition to intelligently recommending objects that may be of interest to the user, the user can also issue an external instruction to select a second object in the scene image. In this way, the processor can combine the marked first target and the second target indicated by the external instruction to perform final target confirmation, and after the target is confirmed, use the confirmed target for target tracking, and make the confirmed target always Stay within the predetermined range of the screen.

Wherein, the final target confirmation is performed by combining the marked target and the target indicated by the external instruction, which may be to confirm that the first target among the selected targets is used for tracking, or to confirm the selected target selected by the user. A second target other than is the target used for tracking. That is, the marked first target and the second target indicated by the external instruction have corresponding priorities, and the priorities may vary according to specific application scenarios. Under normal circumstances, the priority of the second target indicated by the user instruction will be higher than the first marked target. For example, when the second target indicated by the user command is different from the marked first target, the first target indicated by the user command will be The second target is the final target for tracking. However, when the confirmed target is lost and the user no longer indicates the second target through external instructions, the first marked target has the highest priority. For example, when the tracked target is lost, the tracked target can automatically Switches to one of the marked first targets. Of course, it can be understood that in some embodiments, if the tracked target is not lost, the priority of the marked first target may also be higher than that of the second target indicated by the external instruction, which can be designed according to needs. , not specifically limited here.

It should be noted that the second goal may be the first goal, may also include the first goal and goals other than the first goal, and may also be a goal different from the first goal.

Compared with the need to rely on the user to confirm the target through the touch screen, the method provided in this embodiment can actively provide the user with the target that the user may be interested in according to the characteristic information of the first target in the scene, and combine the external command instruction The second goal can finally be confirmed according to the first goal and the second goal. The method provides a more intelligent target confirmation method for the user, liberates the hands of the user in some cases, and improves the user experience of the user.

In an embodiment, the feature information of the first object in the scene image may be various, such as the depth of field of the object, the size of the object, the distance between the object on the screen and the preset position of the scene image, or one or more of the category of the object. Various combinations.

Optionally, the feature information includes the distance between the target on the screen and the preset position of the scene image. Exemplarily, according to the user's photographing habits in general, the user likes to place the most interesting object in the middle of the scene image, so in order to more intelligently recommend the user may be interested in the object, the distance between the object in the scene image can be The center distance of the scene image is used as the feature information of the target. Of course, the preset position of the scene image includes but not limited to the center of the scene image. Assume that the preset position of the scene image is set as the center of the scene image, if there are three targets A, B, and C in the scene image, and the center of the scene image is O, the distance between targets A and O can be expressed as AO, and the distance between targets B and O The distance can be expressed as BO, the distance between target C and O can be expressed as CO, if AO>BO>CO, and the target closest to O is selected as the target for the user, then the processor can select target C as the first target recommended to users. It should be understood that if the preset position of the scene image is set to a position other than the center of the scene image, the principle of selecting the target by the processor is similar to the above method.

Optionally, in some scenes, the user needs to track and shoot a specific target. For example, when studying the living habits of a dog, it is necessary to observe the movement track of a dog. The user hopes that the processor can directly capture the dog in the scene image. As the recommended target is recommended to the user, the processor may also use the category of the target as the reference feature information of the selected target. At the same time, it should be noted that the category of the target is not just animals, plants, human beings and other categories with large differences. The category of the target can also be men, women, old people, young people, children, etc., and the processor can The demand scenario recommends different targets that the user may be interested in for the user. For example, there are three targets D, E, and F in the scene image. After target recognition, the D target is a human, the E target is a cat, and the F target is a dog. If the user demand received by the processor is to pay more attention to the human target, then the processor may select target D as the first recommended target to the user. For another example, there are three targets D1, E1, and F1 in another scene image. After target recognition, D1 is an elderly person, E1 is a young person, and F1 is a child. If the user request received by the processor at this time is a monitoring scene If there is a child in the image, the processor may select the target F1 as the first recommended target to the user.

Optionally, in some specific scenarios, objects that may be of interest to the user may be recommended to the user based on the depth information of the object. Depth of field refers to the relatively clear imaging range before and after the focus of the camera. The depth of field of the target can be understood as the distance between the target and the imaging lens in the actual scene, that is, the farther the target is from the imaging lens, the greater the depth of field of the target, and the closer the target is to the imaging lens. The depth of field is smaller. Exemplarily, if the request received by the processor is to take pictures of distant landscapes, then when there are mountain objects with a large depth of field and human objects with a small depth of field in the scene image, the processor may select the mountain object with a large depth of field as the first choice The target is recommended to the user.

Optionally, in yet another example, the processor may recommend objects that may be of interest to the user according to the size of the objects. For example, in the scene of tracking shooting, the user is interested in the more obvious objects in the scene image, that is, the objects occupying a larger area of the scene image, so compared with other objects, the objects occupying a larger area of the scene image are obviously is the more obvious target. After the processor receives the request, it can recommend the object occupying the largest area of the scene image to the user as the first selection target.

Of course, the characteristic information of the target may not only be a single piece of information, but also may be a superposition of multiple characteristic information. The distance between the target and the center of the scene image in the category superimposed scene image based on the feature information of the target is taken as an example for illustration. For example, in a specific scene, such as tracking and shooting a dance performance, people may be more willing to pay attention to the lead dancer target in the center of the scene image, then obviously the category information of the target is the most basic information, and other feature information is based on the category information Therefore, the processor can give the category information of the target a weight score of 90% (it should be understood that the proportion of the weight score can be adjusted), that is, if the target in the scene image is recognized as a human being, then the A target can get a weight score of 90%, and if the target that recognizes the scene image is a scene set, etc., the target cannot get a 90% weight score, or the target can only get less points. Further, give the feature information of the distance between the object in the scene image and the center of the scene image a weight score of 10%, sort the objects in the scene image according to the distance from the center of the scene image, the objects with small distances are ranked first, and the objects with large distances are ranked first. Ranked in the back, the first goal can get all 10% of the weighted score, and the scores that can be obtained by the last goal are decreasing in turn. Finally, add the score obtained by the target in the category information part and the score obtained in the distance from the center of the scene image to get the final score of the target, and finally recommend the target with the highest final score for the user, or the processor can recommend the target with the highest final score for the user. A series of high goals, and finally confirmed by the user. It should be understood that, except for the specific examples cited in the example, different weight scores are given to different feature information of the target, and finally the method of target selection based on the superimposed total score of all feature information of the target is also applicable to two or more The combination of other feature information of the target.

It can be understood that by concretizing the feature information of the target in the scene image, the target that the user may be more interested in can be more accurately recommended to the user in different application scenarios. The categories and combinations of feature information may be preset in advance or customized by the user, which are not specifically limited here.

It should be noted that the recommended target is determined and pushed by other electronic devices, or calculated by a processor based on feature information, which is not specifically limited here.

In yet another embodiment, there may be various marking methods for marking the target, such as a line marking method and a target area marking method.

Optionally, the line marking method may include using straight lines, curved lines or a combination thereof to mark at least one position around the selected target, such as marking the target with a straight line or wavy line that is different from the color of the screen below the target, and for example Add a straight line to the upper left of the target and a wavy line to the lower right for target marking.

Optionally, the method for marking the target area includes one of delineating the selected target area and adjusting the brightness of the selected target area, that is, marking a specific area in a closed manner. For example, use a rectangle, circle or triangle that is different from the color of the screen to delineate the recommended target. For another example, the overall brightness of the selected target area may be higher than other areas of the scene image. Wherein, the target area can be understood as but not limited to the area indicated by the target outline extracted in the target recognition algorithm. It can be understood that marking the selected target can make the selected target different from other targets in the scene image, and specific marking methods such as star-shaped and heart-shaped marking methods can also make the human-computer interaction interface on the screen more beautiful.

In another embodiment, the external instruction may be an instruction of contact with the screen, an instruction of non-contact with the screen, or a combination of an instruction of contact with the screen and an instruction of non-contact with the screen. The instruction of touching the screen refers to an instruction issued by the user through the touch screen, for example, the user may select a second object by framing or circling the object on the screen. A non-contact instruction with the screen refers to an instruction issued by the user through a non-touch screen, which includes at least one of voice instructions, gesture instructions, and instructions based on operating elements. Exemplarily, the user may select the second target by issuing a voice command such as "left, right" to switch the logo marking the target to be confirmed among the selected targets. Another example is that the user can use a gesture command to select a second target. The gesture command here can refer to a gesture command issued by a certain part of the user's body. For example, the user's head or hands can make a specific gesture to mark the target to be confirmed. The identification of the switch in the recommended target can also refer to the posture of No. 1 hand-held part 11 and/or No. 2 hand-held part 22 as shown in Figure 1 (if No. 1 hand-held part 11 and/or No. 2 hand-held part 22 can Rotate relative to the pan-tilt main body 13), so as to select the second target. For another example, the user can control the operating element set apart from the screen to select the second target. For example, the user can switch the logo of the target to be confirmed among the recommended targets through the button or knob set apart from the screen, so as to select the second target. Two goals. It can be understood that when the method of issuing a non-contact command with the screen to confirm the target is applied to a handheld system with a large mass and is inconvenient to hold with one hand, it can avoid the need for the user to free up one hand to touch the screen to select the target band. It's not convenient to come. And when the target selected by the processor for the user does not include the target that the user is interested in, the user can also issue an instruction to touch the screen to select the target that the user is interested in, so that the final confirmed target can be expressed more accurately.

Among them, the method in the embodiment of the present application can be applied to electronic devices that can be held in both hands. When the electronic device is held in both hands, the operating part that issues the operating part command is located on the electronic device where the user's fingers can operate, and the screen is located on the electronic device. The area where the user's fingers cannot operate. It can be understood that the above-mentioned electronic devices can be diverse, such as two hand-held pan-tilts, remote control devices (such as remote controls, equipped with mobile phone with control software). Exemplarily, the electronic device is a two-handed pan-tilt, and the two-handed pan-tilt is generally a large-scale handheld pan-tilt with a certain weight, which can be used for filming and filming. The side, and the screen for the user to monitor the screen is generally located above the main body of the gimbal, where the user's eyes can directly observe. When in use, the user holds the two-handed pan-tilt with both hands, and when a target of interest appears in the scene image on the screen, the user can obtain some targets recommended according to the feature information of the target in the scene image. In the recommended part of the target, the user can select the target for tracking and shooting by controlling the operating elements (such as buttons and knobs) located in the surrounding area of the two-handed pan-tilt for the user to hold. In this embodiment, this technical solution can not only bring a novel target confirmation method to the user, but also the user does not need to free up a hand to select the target for tracking and shooting on the screen through touch, thus solving the problem of selecting and tracking the target. Inconvenience caused by holding the above-mentioned cloud platform with one hand when aiming. In yet another embodiment, the recommended target may be one target, or multiple targets, which may be specifically designed according to needs.

Exemplarily, there is one selected target, and the selected target is a target at the first preset ranking position among multiple targets, and the identifier used to mark the target at the first preset ranking position is the first specific mark, and multiple The sorting of objects is determined according to the characteristic information of the objects. That is, when the selected target is a target, the first target can be sorted according to the characteristic information of the first target, and the first target at the first preset sorting position can be recommended to the user, and the selected target can be sorted with the first specific mark. In order to distinguish it from other objects in the scene image, the first preset bit can be the highest bit or any preset bit. Wherein, the first specific mark may be a mark described in the target marking method mentioned in the above embodiment, for example, a red line is used to frame the target area, or a wavy line is used to mark the target area. For example, the distance between the first object in the scene image and the center of the scene image is used as the feature information of the first object to sort the first objects. The shorter the distance, the higher the order of the first object. If the sort bit is used as the first preset sort bit, then the processor can select and mark the target with the shortest distance from the center of the scene image among the first targets for the user and mark it with the first specific mark. This selection method can straight-forwardly select the target that the user is most likely to be interested in, and give the user a clear indication of the target.

Optionally, for example, when there are multiple selected targets, at least two selected targets can be marked to obtain marked targets, wherein the marking method of the target at the first preset ranking position is the same as that at the non-th The marking methods of objects at a preset ranking position are different. The mark used to mark the target at the first preset ranking position is the first specific mark, and the ranking of multiple targets is determined according to the feature information of the targets. Wherein, the first specific mark can be the mark described in the target notation method mentioned in the above embodiment, but the first specific mark can be distinguished from other marks, for example, the target of the non-first preset sorting position is framed by a green line Mark the target area, and the first specific mark is to frame the target area with a red line, or the target that is not the first preset sort position is to use a green line to frame the target area mark, and the first specific mark is to frame the target area with a green line Add a star marker to the upper left. For example, the processor may select and mark a plurality of selected targets within a preset sorting range, and may use a first specific flag to mark a target at a first preset sorting position, wherein the preset sorting range consists of the first preset A plurality of sorting bits including the sorting bit is formed. Exemplarily, the distance between the first object in the scene image and the center of the scene image is used as the feature information of the first object to sort the first objects, the shorter the distance, the higher the order of the first object, the default sorting The range is set to the top five sorted targets, and the first preset sort bit is set to the first sorted target. Then the processor can recommend the top five goals for the user, wherein the goal ranked first can be circled by a red line, and the goals ranked second to fourth can be circled by a green line. Using different tags to mark the goals in the first ranking position can more clearly show the user a series of goals that the processor may be interested in and the most likely goals that the processor recommends for the user in this series of goals. A target of interest. Such a selection method can bring more choices to users and improve user experience.

It can be understood that the objects not in the first preset sorting position also include the objects outside the preset sorting range, such as the second target different from the first target indicated by the external instruction.

In yet another embodiment, at least two recommended targets are marked, and after the marked targets are obtained, the user is prompted to confirm the targets by using an instruction of touching the screen. Exemplarily, the processor may select five targets for the user, wherein the target at the first ranking position is marked with the first specific mark, and the remaining four targets are marked with non-first specific marks. After marking the targets, the processor may pass Displaying text on the screen or prompting the user through sound may select the second target by using an instruction of contact with the screen (for example, a way of selecting a target). This method can prompt the user to take the next action after getting the marked target, and in case the marked target does not include the target that the user is really interested in, it can guide the user to perform the corresponding operation to select the target that the user is really interested in, increasing user experience.

In yet another embodiment, the first specific mark of the target at the first preset sorting position can be transferred to the target not at the first preset sorting position, and the target corresponding to the first specific mark is the target to be confirmed. Wherein, the above-mentioned targets not in the first preset sorting position may be targets located in the non-first preset sorting position among the marked targets, such as targets in the second preset sorting position or other sorting positions, or it may be a target set by the user through A target other than the marked target obtained by the command to make contact with the screen.

Optionally, there may be one marked target, and the processor may transfer the first specific mark to a target other than the marked target according to an external instruction. For example, when there is only one marked target, the processor can set the first preset sort position to the first target according to an instruction sent by the user through the touch screen, such as touching the target on the screen or selecting a frame on the screen. The specific mark is transferred to the target selected by the user through the touch screen, and then the target selected by the user through the touch screen is used as a new target to be confirmed.

Optionally, when there are multiple marked targets, the processor may transfer the first specific mark to the corresponding target based on an external instruction, wherein the corresponding target may be one of the marked targets or a marked target on a target other than the following target, such as a target selected by the user with an on-screen touch. Exemplarily, when the marked target is a plurality of targets, among which the target in the first preset sort position has the first specific mark, the user can issue an external instruction to transfer the first specific mark to the marked target Targets located in other sort positions, the user can also issue an external command to contact the screen to select a target other than the target after the mark, such as selecting a target other than the target after the mark on the screen to transfer the first specific mark to On a target other than the marked target above. In this method, the first specific mark indicates the target to be confirmed, and the user can control the transfer of the first specific mark, which means that the user can select the target to be confirmed and specify which target is currently selected to be confirmed. Improve user experience for users.

In one embodiment, the corresponding object includes an object selected by touch on the screen. In the above-mentioned embodiments, the case where the corresponding object includes the object selected by touch on the screen has been explained.

It can be understood that the target of the first preset ranking position may be the target that the machine selects to be most likely to be of interest to the user, and the first specific mark is used to mark the target of the first preset ranking position, so that the first preset ranking position The goal of the bit is the goal that the user first needs to be confirmed. However, in some cases, the target that the user is interested in may not be the target of the first preset ranking position, so that the first specific mark of the target to be confirmed can be switched among different targets, which can better The user is prompted which target is currently to be confirmed, and at the same time, the first specific mark is reused to indicate the target that the user may be most interested in, especially after the user actively selects a target other than the target in the first preset ranking position.

It should be noted that the target to be confirmed can be marked with an identifier different from the first specific mark, so that in any case, the target of the first preset sort can be clearly indicated, and it is also clear which target to be confirmed is.

In yet another embodiment, the confirmed target may have a second specific mark, the second specific mark mentioned may be the mark made by the above-mentioned marking method, and the second specific mark may be different from the non-first preset sorting The mark of the target of the position and the mark of the first specific mark, for example, the mark of the target of the non-first preset sorting position is framed by a green line outside the target, the first specific mark is framed by a red line outside the target, and the second specific mark To add a star marker in the target area. Certainly, the second specific mark may also be the same as the first specific mark, for example, the confirmed target is the target marked by the first specific mark.

In yet another embodiment, if the currently confirmed target is lost and the currently confirmed target is the target at the first preset ranking position, the first specific mark is used to mark the target at the second preset ranking position. The method provided in this embodiment describes another transfer rule of the first specific flag when the currently confirmed target is lost. Exemplarily, the currently confirmed target may be the target being tracked and photographed, and the currently confirmed target is the target at the first preset sort position when sorting the targets in the scene image, if the target is lost, Then use the first specific mark to mark the target located in the second preset sorting position in the original scene image, such as using the feature information of No. 1, 2, and 3 targets in the scene image to sort No. 1, 2, and 3 targets , the sequence from front to back is No. 1 target, No. 3 target, and No. 2 target. If No. 1 target is the target with the first specific mark and is currently confirmed, then after No. 1 target is lost, the No. 1 target can be used A specific mark marks the No. 3 target, that is, the No. 3 target is a reconfirmed target at the first preset sorting position. In this way, it can be determined which target is always in the first preset sorting position, and also improves the transfer method of the first specific mark when the tracking target is lost in different scenarios.

In yet another embodiment, performing target confirmation based on the external instruction and the marked target includes: determining one of the marked targets as the currently confirmed target based on the external command. The user can issue an external command not in contact with the screen or an external command in contact with the screen to select a target among the marked targets as the final confirmed target for target tracking. Exemplarily, when target confirmation is performed in the application scenario of a handheld pan/tilt, after selecting and marking at least one target for the user according to the feature information of the target, the user can select among the marked targets by issuing key instructions such as A target as the currently confirmed target. It is understandable that selecting a currently confirmed target for tracking from the marked targets makes the target confirmation process more intelligent. At the same time, in some cases, the user's hands can be freed, that is, the user can operate without leaving the finger The target area can be confirmed, which brings a certain degree of convenience to the user.

In yet another embodiment, performing target confirmation based on the external instruction and the marked target includes: determining a target other than the marked target as the currently confirmed target based on the external command. The user can issue an external command to touch the screen and touch on the screen to select a target other than the marked target as the currently confirmed target. Exemplarily, when performing target confirmation in the application scenario of the handheld pan/tilt, after selecting and marking at least one target for the user according to the characteristic information of the target, the user can select another target by frame-selecting the target on the screen , and make that target the current confirmed target. It should be understood that when the target confirmation method is applied to the target selected and marked for the user is not the target that the user is really interested in, the user can select the target that the user is really interested in in the scene image as the target to track. To a certain extent, this method can correct some deviations caused by the intelligent selection algorithm, and can allow users to confirm targets more accurately.

In yet another embodiment, if the currently confirmed target is lost, it is determined that one of the marked targets is the currently reconfirmed target. Exemplarily, if the currently confirmed target is one of the marked targets, and when the target is lost, the target in the scene image at this time can be acquired (because the target is lost, the scene image at this time and the confirmed The scene image at the time of the target is not the same), and the scene image target at this time is sorted according to the feature information of the target, and the target at the first preset sorting position is selected, marked and determined as the current reconfirmation , and then switch to target tracking on the currently redefined target. Or, for example, the currently confirmed target is a target other than the marked target, and when the target is lost, the target in the scene image at this time can also be obtained, and the target at this time can be determined according to the characteristic information of the target Sorting is performed, and the target at the first preset sorting position at this time is selected, marked and determined as the current reconfirmed target. The method provided in this embodiment provides the user with a method when the tracked target is lost, so that the user does not need to go through the process of confirming the target again when the target is lost, which brings certain convenience to the user , improving the user experience.

In yet another embodiment, if the currently confirmed target is retrieved, one of the marked targets is switched back to the currently confirmed target. That is, after the lost target is recovered, the tracked object is switched back to the recovered lost target. For example, the target H is the currently confirmed target for tracking and shooting. When the target H is lost, the tracking object will be switched to the target I selected and marked in the current scene image. If the target H is in the current scene image is retrieved, the tracking object is switched from target I to target H. It can be understood that the process of recovering the target is based on the next improvement of the process of losing the target. Usually, when the user does not manually stop the target tracking process, the lost target is often the target that the user is most interested in. This target If it appears in the scene image after being lost, the user will probably want to continue to track the retrieved object. The method provided by this embodiment solves the problem that the user needs to manually re-determine the retrieved object as the tracking object. inconvenient problem.

The target confirmation method in the embodiment of the present invention has been described above, and the target confirmation device in the embodiment of the present invention will be described below:

FIG. 3 is a schematic structural diagram of a target confirmation device provided by an embodiment of the present application. As shown in FIG. 3 , the target confirmation device 30 may include: a memory 31 , a processor 32 and a screen 33 .

A screen 33 for displaying scene images and marked targets;

Memory 31, used to store computer programs that can run on the processor 32;

The processor 32 is configured to perform the following operations when running the computer program:

Acquiring the recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Mark the selected target to obtain the marked target;

acquiring an external instruction, and performing target confirmation based on the external instruction and the marked target, where the external instruction is used to select a second target in the scene image;

The confirmed target is used for target tracking, so that the confirmed target remains within a predetermined range of the screen, and the screen is used to display the scene image and the marked target.

The target confirmation device provided in this embodiment can be used to implement the technical solutions of all the above-mentioned method embodiments, and its implementation principle and technical effect are similar to those of the method embodiments, and will not be repeated here.

In addition, the embodiment of the present application also provides a handheld system 40 as shown in FIG. 4, including a handheld device 41, a memory 42, a processor 43, and a screen 44. Handheld device 41 communication connection;

The hand-held device 41 includes a hand-held part, and the hand-held part is used for being held by the user;

Memory 42, used for storing the computer program that can run on processor 43;

The processor 43 is configured to perform the following operations when running the computer program:

Mark the selected target to obtain the marked target;

Obtaining an external instruction, and performing target confirmation based on the external instruction and the marked target, the external instruction is used to select a second target in the scene image;

Use the confirmed target to track the target so that the confirmed target remains within the predetermined range of the screen;

a screen for displaying the scene image and the marked target.

The handheld system provided in this embodiment can be used to implement the technical solutions of all the above-mentioned method embodiments, and its implementation principle and technical effect are similar to those of the method embodiments, and will not be repeated here. Optionally, the screen is detachably connected to the handheld device. Exemplarily, when the handheld system is a mobile phone stabilizer, when the mobile phone stabilizer is working, the mobile phone will generally be fixed on the mobile phone stabilizer to play the role of the screen. However, if there is an incoming call or the user needs to touch the screen to control the mobile phone, the user can also disassemble the mobile phone and use it.

In an embodiment, the handheld device includes one of a handheld pan-tilt and a control device; wherein the control device is used to control a movable platform. It can be understood that the form of the handheld device can be various, for example, the handheld device can be a hand-held pan-tilt (hand-held pan-tilt can include a mobile phone stabilized pan-tilt, a camera-stabilized pan-tilt) or a mobile platform (such as an unmanned platform) machine, unmanned vehicle, unmanned ship), such as a remote control to control a movable platform or a mobile phone equipped with software to control a movable platform. The target confirmation methods mentioned above can all be applied to the handheld system with the above-mentioned handheld device, and bring corresponding beneficial effects.

In yet another embodiment, the handheld system further includes an image acquisition device, which is installed on the handheld device or remotely communicated with the handheld device; the image acquisition device is used to acquire scene images.

The hand-held part of the handheld device provided in this embodiment is used for the user to hold when operating the hand-held system. During the operation of the hand-held system, the hand-held part can be integrated with the image acquisition device, or can be integrated with the image acquisition device are physically separated.

Exemplarily, when the handheld system is working, the image acquisition device and the handheld component are physically integrated, such as a camera stabilizer, a mobile phone stabilizer, and an electronic device in which the camera and the stabilizer are integrated. At this time, the user can obtain the target that the user may be interested in selected and marked by the above-mentioned electronic device, the user sends an external instruction to confirm the target of interest, and the processor uses the confirmed target for target tracking.

Or, exemplary, the handheld system is a system in which a mobile platform such as an unmanned aerial vehicle, an unmanned vehicle, or an unmanned ship is matched with a control device. When the handheld system is working, the image acquisition device and the control device are physically separated. The user can hold the handheld device to control the movable platform. The shape of the handheld device can be diverse, such as the above-mentioned unmanned vehicle, The remote controller matched with the unmanned ship or the mobile phone equipped with control software matched with the above-mentioned unmanned vehicle and unmanned ship. Exemplarily, the handheld system is a system composed of a drone equipped with an image acquisition device and its supporting remote control. When the system is working, the image acquisition device and the remote control are separated. The target confirmation method mentioned above can be applied to this system.

It should be noted that the present application also provides a computer program, which can be used to implement any one of the above methods when the computer program is executed by a computer device. The computer device may be a computing device such as a desktop computer, a notebook, a palmtop computer, or a cloud server. A computer device may include, but is not limited to, a processor, memory. Those skilled in the art can understand that the processor and memory are just examples of computer devices, and do not constitute a limitation to computer devices, and may include more or less components, or combine certain components, or different components, such as computer The device may also include input and output devices, network access devices, buses, and so on.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by program instructions and related hardware. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Claims

A target confirmation method, characterized in that, comprising:

Acquiring a recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Marking the selected target to obtain the marked target;

Acquiring an external instruction, and performing target confirmation based on the external instruction and the marked target, the external instruction is used to select a second target in the scene image;

The confirmed target is used for target tracking, so that the confirmed target remains within a predetermined range of a screen for displaying the scene image and the marked target.
The method according to claim 1, wherein the external instruction comprises: an instruction of contact with the screen and/or an instruction of non-contact with the screen.
The method according to claim 2, wherein the non-contact instructions with the screen include one of voice instructions, gesture instructions, and instructions based on operating elements, and the operating elements are arranged separately from the screen. .
The method according to claim 3, characterized in that the method is applied to an electronic device that can be held by both hands, and when the electronic device is held by both hands, the operating member that issues the operating member instruction is located on the electronic device. The area where the fingers of the user can operate, and the screen is located on the electronic device where the user's fingers cannot be operated.
The method according to claim 1, wherein the feature information of the target includes the depth of field of the target, the size of the target, and the distance between the target and the preset position of the scene image on the screen. At least one of distance and category of the target.
The method according to any one of claims 1 to 5, characterized in that there is one selected target, and the selected target is a target located in the first preset sorting position among a plurality of targets, which is used for marking The identifier of the object at the first preset ranking position is a first specific mark, and the ranking of the multiple objects is determined according to the feature information of the objects.
The method according to any one of claims 1 to 5, wherein there are multiple selected targets, and marking the selected targets to obtain marked targets includes:

Mark at least two of the selected targets to obtain marked targets, wherein the marking method of the target at the first preset ranking position is different from the marking method of the target at the non-first preset ranking position, and is used for marking The identifier of the object at the first preset ranking position is a first specific mark, and the ranking of the multiple objects is determined according to the feature information of the objects.
The method according to claim 7, characterized in that, after marking at least two of the selected targets and obtaining the marked targets, the method further comprises:

Prompting the user to enable target confirmation with an instruction to touch the screen.
The method according to claim 6 or 7, characterized in that the first specific mark of the object located in the first preset sort position can be transferred to the target located in the non-first preset sort position, wherein the The target corresponding to the first specific mark is the target to be confirmed.
The method according to claim 9, characterized in that the method further comprises:

Transferring the first specific token to a corresponding target based on the external instruction.
The method according to claim 10, wherein the corresponding object comprises an object selected by touch on the screen.
The method according to any one of claims 6 to 11, further comprising:

If the currently confirmed target is lost and the currently confirmed target is the target at the first preset ranking position, the first specific mark is used to mark the target at the second preset ranking position.
The method according to claim 1, wherein the target confirmation based on the external instruction and the marked target comprises:

Based on the external instruction, it is determined that one of the marked targets is the currently confirmed target.
The method according to claim 1, wherein the target confirmation based on the external instruction and the marked target comprises:

Based on the external instruction, it is determined that a target other than the marked target is the currently confirmed target.
The method according to claim 13 or 14, wherein the method further comprises:

If the currently confirmed target is lost, it is determined that one of the marked targets is the currently reconfirmed target.
The method according to claim 15, further comprising:

If the currently confirmed target is retrieved, one of the marked targets is switched back to the currently confirmed target.
The method according to claim 13 or 14, characterized in that the confirmed target has a second specific mark.
The method according to claim 1, wherein the marking method for marking the proposed target comprises: at least one of a line marking method and a target area marking method.
The method according to claim 18, characterized in that, the shape of the line in the line marking method includes at least one of a straight line and a curve;

The method for marking the target area includes at least one of delineating the target area and adjusting brightness changes of the target area.
A target confirmation device, characterized in that the device includes: a memory, a processor, and a screen;

said memory for storing a computer program capable of running on said processor;

The processor is configured to perform the following operations when running the computer program:

Acquiring a recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Marking the selected target to obtain the marked target;

Acquiring an external instruction, and performing target confirmation based on the external instruction and the marked target, the external instruction is used to select a second target in the scene image;

performing target tracking by using the confirmed target, so that the confirmed target remains within a predetermined range of the screen;

The screen is used to display the scene image and the marked target.
The device according to claim 20, wherein the external instruction comprises: an instruction of contacting the screen and/or an instruction of non-contacting the screen.
The device according to claim 21, wherein the non-contact instruction includes one of a voice instruction, a gesture instruction, and an instruction based on an operating element, and the operating element is provided separately from the screen.
The device according to claim 22, characterized in that the device is applied to an electronic device that can be held by both hands, and when the electronic device is held by both hands, the operating part that issues the instruction of the operating part is located on the electronic device. The area where the fingers of the user can operate, the screen is located on the electronic device where the user's fingers cannot be operated, and the device is a part of the electronic device.
The device according to claim 20, wherein the characteristic information of the target includes the depth of field of the target, the size of the target, the distance between the target and the preset position of the scene image, the target at least one of the categories.
The device according to any one of claims 20 to 24, characterized in that there is one selected target, and the selected target is a target in the first preset ranking position among multiple targets, which is used for marking The identifier of the object at the first preset ranking position is a first specific mark, and the ranking of the multiple objects is determined according to the feature information of the objects.
The device according to any one of claims 20 to 24, wherein there are multiple selected targets, and the processor is used to mark the selected targets to obtain marked targets, specifically comprising:

Mark at least two of the selected targets to obtain the marked targets, wherein the marking method of the target at the first preset ranking position is different from the marking method of the target at the non-first preset ranking position, and is used for marking The identifier of the object at the first preset ranking position is a first specific mark, and the ranking of the multiple objects is determined according to the feature information of the objects.
The device according to claim 26, wherein, after the processor is configured to mark at least two of the selected targets, and after obtaining the marked targets, the processor is further configured to:

Prompting the user to enable target confirmation with an instruction to touch the screen.
The device according to claim 25 or 26, wherein the first specific mark of the object located in the first preset sort position can be transferred to the target located in the non-first preset sort position, wherein the The target corresponding to the first specific mark is the target to be confirmed.
The device according to claim 28, wherein the processor is further configured to:

Transferring the first specific token to a corresponding target based on the external instruction.
The apparatus according to claim 29, wherein the corresponding object comprises an object selected by touch on the screen.
The device according to any one of claims 25 to 30, wherein the processor is further configured to:

If the currently confirmed target is lost and the currently confirmed target is the target at the first preset ranking position, the first specific mark is used to mark the target at the second preset ranking position.
The device according to claim 20, wherein the processor is configured to perform target confirmation based on the external instruction and the marked target, specifically comprising:

Based on the external instruction, it is determined that one of the marked targets is the currently confirmed target.
The device according to claim 20, wherein the processor is configured to perform target confirmation based on the external instruction and the marked target, specifically comprising:

Based on the external instruction, it is determined that a target other than the marked target is the currently confirmed target.
The device according to claim 32 or 33, wherein the processor is further configured to:

If the currently confirmed target is lost, it is determined that one of the marked targets is the currently reconfirmed target.
The device according to claim 34, wherein the processor is further configured to:

If the currently confirmed target is retrieved, one of the marked targets is switched back to the currently confirmed target.
The device according to claim 32 or 33, wherein the confirmed object has a second specific mark.
The device according to claim 20, wherein the marking method for marking the proposed target comprises: at least one of a line marking method and a target area marking method.
The device according to claim 37, wherein the shape of the line in the line notation method includes at least one of a straight line and a curve;

The method of marking the target area includes at least one of brightness change of the target area and adding markers to the target area.
A handheld system, characterized in that it includes: a handheld device, a memory, a processor, and a screen, the memory and the processor are set on the handheld device, and the screen is connected in communication with the handheld device;

The hand-held device includes a hand-held part for being held by a user;

said memory for storing a computer program capable of running on said processor;

The processor is configured to perform the following operations when running the computer program:

Acquiring a recommended target, wherein the selected target is determined according to the feature information of the first target in the scene image;

Marking the selected target to obtain the marked target;

Acquiring an external instruction, and performing target confirmation based on the external instruction and the marked target, the external instruction is used to select a second target in the scene image;

performing target tracking by using the confirmed target, so that the confirmed target remains within a predetermined range of the screen;

The screen is used to display the scene image and the marked target.
The handheld system according to claim 39, wherein the external instruction includes: an instruction of contact with the screen and/or an instruction of non-contact with the screen.
The handheld system according to claim 40, wherein the non-contact instructions with the screen include one of voice instructions, gesture instructions, and instructions based on operating elements, and the operating elements are arranged separately from the screen. .
The handheld system according to claim 41, characterized in that, when the handheld device is held in both hands, the operating part that issues the operating part command is located on the handheld device where the user's fingers can be operated, and the screen is located at An area of the hand-held device that cannot be manipulated by the user's fingers.
The handheld system according to claim 39, wherein the feature information of the target includes the depth of field of the target, the size of the target, the distance between the target and the center point of the scene image, the target at least one of the categories.
According to the handheld system according to any one of claims 39 to 43, it is characterized in that there is one selected target, and the selected target is a target located in the first preset sorting position among multiple targets, and multiple The sorting of the objects is determined according to the feature information of the objects.
The handheld system according to any one of claims 39 to 43, wherein there are multiple selected targets, and the processor is used to mark the selected targets to obtain marked targets, specifically comprising:

Mark at least two of the selected targets to obtain the marked targets, wherein the marking method of the target at the first preset ranking position is different from the marking method of the target at the non-first preset ranking position, and multiple selected targets The ranking of the objects is determined according to the characteristic information of the objects.
The handheld system according to claim 45, wherein the processor is used to mark at least two of the selected targets, and after obtaining the marked targets, the processor is also used to prompt the user to use the The command to contact carries out target confirmation.
The hand-held system according to claim 44 or 45, wherein the processor is further configured to transfer the first specific mark of the object located at the first preset sorting position to a position other than the first preset sorting position. on the target, wherein the target corresponding to the first specific mark is the target to be confirmed.
The handheld system according to claim 47, wherein the processor is further configured to transfer the first specific marker to a corresponding target based on the external command.
The handheld system of claim 48, wherein said corresponding object comprises an object selected by touch on said screen.
The handheld system according to any one of claims 47 to 49, wherein the processor is further used for:

If the currently confirmed target is lost and the currently confirmed target is the target at the first preset ranking position, the first specific mark is used to mark the target at the second preset ranking position.
The handheld system according to claim 39, wherein the processor is configured to perform target confirmation based on the external instruction and the marked target, specifically comprising:

Based on the external instruction, it is determined that one of the marked targets is the currently confirmed target.
The handheld system according to claim 39, wherein the processor is configured to perform target confirmation based on the external instruction and the marked target, specifically comprising:

Based on the external instruction, it is determined that a target other than the marked target is the currently confirmed target.
The handheld system according to claim 51 or 52, wherein the processor is also used for:

If the currently confirmed target is lost, it is determined that one of the marked targets is the currently reconfirmed target.
The handheld system according to claim 53, wherein the processor is further used for:

If the currently confirmed target is retrieved, one of the marked targets is switched back to the currently confirmed target.
The hand-held system according to claim 51 or 52, characterized in that the identified object has a second specific mark.
The hand-held system according to claim 39, wherein the marking method for marking the proposed target includes at least one of a line marking method and a target area marking method.
The handheld system according to claim 56, characterized in that, the shape of the line in the line notation method includes at least one of a straight line and a curve;

The method of marking the target area includes at least one of brightness change of the target area and adding markers to the target area.
The handheld system according to claim 39, wherein the screen is detachably connected to the handheld device.
The handheld system according to claim 39, wherein the handheld device includes one of a handheld pan-tilt and a control device;

Wherein, the control device is used to control the movable platform.
The handheld system according to claim 39, characterized in that, the handheld system further comprises an image acquisition device, and the image acquisition device is installed on the handheld device or remotely connected with the handheld device;

The image acquisition device is used to acquire the scene image.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, the computer program includes at least one piece of code, and the at least one piece of code can be executed by a computer to control the computer to perform the claims The method described in any one of 1-19.
A computer program, characterized in that, when the computer program is executed by a computer device, it is used to realize the method described in any one of claims 1-19.