Anchoring positioning method, apparatus, device and medium

Abstract

The application discloses an anchor point positioning method, device, equipment and medium, and the method comprises the following steps: obtaining the pixel coordinates of a target feature point in an RGB image; determining the detection area corresponding to the target feature point in the RGB image based on the pixel coordinates; and calculating the anchor point coordinates corresponding to the target feature point in the detection area in the RGB image based on a preset algorithm. According to the embodiment of the application, the detection area corresponding to the target feature point is set, and then the preset algorithm is used to calculate the accurate anchor point position corresponding to the target feature point in the detection area, so that accurate positioning of the anchor point is realized, and reliable basis is provided for subsequent ranging, target positioning and other applications.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the technical field of image processing, and particularly relates to an anchor point positioning method and device, equipment and a medium. BACKGROUND

[0002] With the development of electronic devices, combined with various sensor fusion and optimization technologies, there has been obvious optimization in scale estimation, environmental three-dimensional perception, anchor point positioning and tracking, etc. For example, through a mobile phone, measurement in a three-dimensional space can be performed, making the measurement task more convenient. To realize measurement in a real scene by means of an electronic device, it is necessary to first find a conversion relationship between a camera coordinate system and a real world coordinate system, that is, given a point in an image, the 3D position of the point in a real scene, i.e., an anchor point position, needs to be calculated, and then measurement operation is performed by means of the determined anchor point position. For the measurement function using an anchor point in space, the accuracy of the anchor point position is very important.

[0003] At present, in a ranging scene, the mapping relationship between an RGB image and a depth image is usually determined based on double-camera calibration, and then the spatial coordinates of the feature points determined by the mapping relationship are used as anchor point coordinates to calculate the straight-line distance between two points; or the RGB image is used alone, anchor point positioning is performed based on an sfm algorithm (Structure From Motion) architecture, and calculation of the spatial distance is realized.

[0004] For the above anchor point positioning process, due to the difference in resolution between the RGB image and the depth image, there is a large error in the mapping relationship; and the use of the sfm algorithm has poor effect in many scenes, and the time cost of the algorithm is high. SUMMARY

[0005] In view of the above defects or deficiencies in the prior art, it is desirable to provide an anchor point positioning method, device, electronic device and medium, which calculates an anchor point corresponding to a target feature point in a detection region by using a preset algorithm, so as to improve the accurate positioning of the anchor point.

[0006] In a first aspect, an embodiment of the present application provides an anchor point positioning method, which comprises:

[0007] obtaining a pixel coordinate of a target feature point in an RGB image;

[0008] determining a detection region corresponding to the target feature point in the RGB image based on the pixel coordinate;

[0009] calculating an anchor point coordinate corresponding to the target feature point in the detection region in the RGB image based on a preset algorithm.

[0010] In some embodiments, the method comprises:

[0011] obtain a pixel coordinate of the target feature point in the depth image and a detection region corresponding to the target feature point in the depth image;

[0012] calculate, based on a preset algorithm, an anchor point coordinate of the target feature point in the detection region in the depth image.

[0013] In some embodiments, the preset algorithm includes a plurality of operators, and calculating, based on the preset algorithm, the anchor point coordinate of the target feature point in the detection region in the RGB image or in the detection region in the depth image includes:

[0014] calculating, based on each operator, a pre-anchor point coordinate corresponding to the target feature point;

[0015] fusing all the pre-anchor point coordinates obtained based on the weight corresponding to each operator to obtain the anchor point coordinate corresponding to the target feature point.

[0016] In some embodiments, the weight is obtained through machine learning training.

[0017] In some embodiments, the obtaining the pixel coordinate of the target feature point in the depth image and the detection region corresponding to the target feature point in the depth image includes:

[0018] obtaining a mapping relationship between the RGB image and the depth image based on dual-camera calibration;

[0019] obtaining, based on the pixel coordinate of the target feature point and the mapping relationship, the pixel coordinate of the target feature point in the depth image and the detection region corresponding to the target feature point in the depth image.

[0020] In some embodiments, the determining the detection region corresponding to the target feature point includes:

[0021] setting a size of the detection region based on an empirical value;

[0022] obtaining the detection region based on the size and the pixel coordinate.

[0023] In a second aspect, an embodiment of the present application provides an anchor point positioning device, which includes:

[0024] an obtaining module configured to obtain a pixel coordinate of a target feature point in an RGB image;

[0025] a defining module configured to determine a detection region of the target feature point in the RGB image based on the pixel coordinate;

[0026] a calculating module configured to calculate, based on a preset algorithm, an anchor point coordinate corresponding to the target feature point in the detection region in the RGB image.

[0027] In some embodiments, the preset algorithm includes a plurality of operators, and the computing module is specifically configured to:

[0028] The preset anchor point coordinates corresponding to the target feature point are calculated based on each operator;

[0029] The obtained all preset anchor point coordinates are fused based on the weight corresponding to each operator to obtain the anchor point coordinates corresponding to the target feature point.

[0030] In a third aspect, an electronic device is provided, which includes a first coil, a second coil, a near field communication module, a wireless charging receiving module, a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The processor is configured to implement the anchor point positioning method of the first aspect.

[0031] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program for implementing the anchor point positioning method of the first aspect.

[0032] The anchor point positioning method, device, equipment and medium provided by the embodiments of the present application set the detection area corresponding to the target feature point, and then calculate the accurate anchor point position corresponding to the target feature point in the detection area by using a preset algorithm, so as to realize accurate positioning of the anchor point and provide a reliable basis for subsequent ranging, target positioning and other applications. BRIEF DESCRIPTION OF DRAWINGS

[0033] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0034] Figure 1 A flowchart of the anchor point positioning method of the embodiments of the present application is shown;

[0035] Figure 2 A flowchart of the anchor point positioning method of some embodiments of the present application is shown;

[0036] Figure 3 A flowchart of the anchor point positioning method of some embodiments of the present application is shown;

[0037] Figure 4 A flowchart of the anchor point positioning method of some embodiments of the present application is shown;

[0038] Figure 5 A flowchart of the anchor point positioning method of some embodiments of the present application is shown;

[0039] Figure 6 A flowchart of the anchor point positioning method of some embodiments of the present application is shown;

[0040] Figure 7 A flowchart of an anchor point positioning method in some embodiments of the present application;

[0041] Figure 8 A structural diagram of an anchor point positioning device in embodiments of the present application;

[0042] Figure 9 A structural diagram of a computer of a processing device in embodiments of the present application. DETAILED DESCRIPTION

[0043] The present application will be further described below in conjunction with the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related disclosure, but not to limit the disclosure. In addition, it should be noted that only parts related to the disclosure are shown in the drawings for ease of description.

[0044] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0045] It can be understood that in actual image applications, such as image processing scenes of target detection, distance measurement and three-dimensional imaging, it is necessary to accurately determine the feature points or anchor points indicated by the user according to the scene requirements to achieve the user's purpose.

[0046] In embodiments of the present application, in order to improve the accuracy of anchor point positioning, a method for positioning anchor points is provided, which sets a region of interest (ROI) corresponding to a specified feature point, and then uses multiple operators to extract and calculate the target anchor point in the region of interest to accurately position the anchor point.

[0047] In order to better understand and illustrate the anchor point positioning method of the embodiments of the present application, the following will be described in detail. Figures 1 to 9

[0048] Figure 1 A flowchart of an anchor point positioning method in embodiments of the present application, as shown in Figure 1 The method can be executed by a computer processing device, and the method specifically can include:

[0049] S110, obtaining a pixel coordinate of a target feature point in an RGB image.

[0050] S120, determining a detection region corresponding to the target feature point in the RGB image based on the pixel coordinate.

[0051] S130, calculating an anchor point coordinate corresponding to the target feature point in the detection region in the RGB image based on a preset algorithm.

[0052] ​Specifically, in the user processing image processing, the RGB image can be operated, such as selecting a target feature point by touch or other means, so that the processing device can obtain the pixel coordinates of the target feature point in response to the operation.

[0053] For example, in the imaging of the RGB image, the boundary point is selected by touch or mouse to input the position of the target feature point to the processing device, and the processing device identifies the pixel coordinates of the feature point in response to the user operation.

[0054] It can be understood that considering the high resolution of the RGB image, the difference in resolution when the processing device displays the image, and the difference in recognition accuracy, the feature point recognized by the processing device, i.e. the pixel coordinates, and the target feature point indicated by the user, i.e. the anchor point position, are different. In order to ensure that the processing device can identify the accurate anchor point position, further processing can be performed on the identified pixel coordinates.

[0055] Specifically, according to the pixel coordinates of the identified target feature point and the pre-set empirical value, a detection region corresponding to the target feature point, i.e. ROI, is set in the RGB image. Then the feature point detection can be performed in the detection region in the RGB image by using a pre-set algorithm to locate the actual anchor point, i.e. to calculate the accurate anchor point coordinate position of the target feature point.

[0056] It can be understood that the size of the detection region, i.e. the window size, can be set to be circular or rectangular according to the actual application, and the target feature point is located in the detection region, which is not limited by the embodiments of the present application.

[0057] For example, the pixel coordinate value of the target feature point A in the RGB image is denoted as A(x, y), and the detection region set according to the empirical value can be Roi_rag_A(x, y, w, h). Finally, the anchor point coordinate of the feature point A can be calculated in the detection region by using a pre-set algorithm.

[0058] Optionally, in the embodiments of the present application, when calculating the anchor point coordinate in the detection region, a plurality of operators can be used to calculate, i.e. each operator calculates its corresponding coordinate value, and then the weight value corresponding to each operator is fused to obtain the final anchor point coordinate value. The specific calculation process is as shown in Figure 2 and Figure 3 .

[0059] As shown in Figure 2 , the calculation process includes:

[0060] S121, calculating the pre-anchor point coordinate corresponding to the target feature point based on each operator.

[0061] S122, fusing all the pre-anchor point coordinates obtained based on the weight corresponding to each operator to obtain the anchor point coordinate corresponding to the target feature point.

[0062] Specifically, after obtaining the detection region of the target feature point, a pre-stored algorithm can be called to calculate the anchor point coordinate.

[0063] It can be understood that a plurality of operators are pre-stored in the processing device for feature point detection to locate the anchor point, such as an edge extraction operator, a foot point extraction operator, and a feature detection operator.

[0064] After one operation of each operator, the anchor point coordinate corresponding to each operator, i.e., the pre-anchor point coordinate, can be obtained, and then the weight W corresponding to each operator pre-set can be used to fuse all the pre-anchor point coordinates obtained to obtain the final anchor point coordinate, i.e., the target anchor point coordinate value.

[0065] For example, as shown in the above RGB image, A1(x, y), A2(x, y), …, and An(x, y) corresponding to operator 1, operator 2, and operator n can be calculated. Then, after multiplying each pre-anchor point coordinate with the corresponding weight and summing, the target anchor point coordinate value can be obtained. Figure 3

[0066] Optionally, in an embodiment, the weight W can be obtained through machine learning training, and the training process is as shown in Figure 4

[0067] Specifically, first, a weight value can be pre-set, and the pixel coordinate value of the feature point, the detection region of the feature point, and the accurate anchor point coordinate value of the feature point are input, and then each operator is used to calculate the anchor point coordinate corresponding to each operator. The weight value pre-set corresponding to each operator is used to fuse to obtain the estimated coordinate value of the anchor point corresponding to the feature point. Finally, the estimated anchor point coordinate value calculated and the accurate anchor point coordinate value are compared to calculate the difference value, i.e., the loss function. Finally, the weight pre-set can be continuously adjusted according to the loss function, so that the loss function obtained finally reaches the pre-set standard to determine the reasonable weight and complete the machine learning training.

[0068] ​​For example, the RGB image input weight W, and the pixel coordinate value A(x, y) of the feature point, the detection region Roi_A(x, y, w, h) of the feature point and the accurate anchor point coordinate value Label_(x, y) of the feature point can be pre-set. Then, a plurality of operators such as operator 1, operator 2 and operator n can be used to calculate in the detection region to obtain the anchor point coordinates A1(x, y), A2(x, y)…An(x, y) corresponding to each operator. Further, the obtained all coordinates are fused by using the pre-set weight to obtain the estimated anchor point coordinate Estimate_A(x, y) of the specific point, and then the difference Δ = |Estimate_A-Label_(x, y) between the estimated anchor point coordinate and the pre-input standard anchor point coordinate is calculated 2 , that is, the loss function; finally, it is judged whether the loss function value is within the pre-set threshold range, if not within the pre-set range, the weight corresponding to the operator is adjusted to adjust the loss function, and finally the loss function value reaches the pre-set standard, and the weight training is completed.

[0069] It can be understood that the weight can also be set according to the experience value, and the embodiments of the present application do not limit this.

[0070] The anchor point positioning method provided by the embodiments of the present application sets the detection region corresponding to the user-specified feature point, and then calculates the accurate anchor point position corresponding to the feature point in the detection region by using a pre-set algorithm, so as to realize accurate calculation of the anchor point and provide a reliable basis for subsequent distance measurement, target positioning and the like.

[0071] Further, in actual application, the RGB image and the depth image are usually combined to realize distance measurement or target positioning purposes. Therefore, on the basis of the anchor positioning of the target feature point in the RGB image, the anchor point corresponding to the target feature point still needs to be positioned in the depth image.

[0072] As shown in Figure 5 , the method can further include:

[0073] S140, obtaining a mapping relationship in the RGB image and the depth image based on the dual-camera calibration;

[0074] S150, obtaining the pixel coordinate of the target feature point in the depth image and the detection region corresponding to the target feature point in the depth image based on the mapping relationship and the pixel coordinate of the target feature point;

[0075] S160, calculating the anchor point coordinate of the target feature point in the detection region in the depth image based on a pre-set algorithm.

[0076] Specifically, in target detection or ranging scenarios, this can be achieved by combining RGB and depth images. During processing, dual-camera calibration is first used to obtain the mapping relationship between feature points in the RGB and depth images. Then, after the processing device obtains the pixel coordinates of the feature points in the RGB image and the defined detection area, the mapping relationship, pixel coordinate values, and detection area can be used to determine the pixel coordinates of the target feature points in the depth image and the corresponding detection area.

[0077] For example, the mapping relationship can be used to obtain the depth image corresponding to feature A in the RGB image. Then feature points are obtained. The detection area, such as

[0078] Similarly, such as Figure 6 As shown, after obtaining the depth detection region, the anchor coordinates of the target feature points can be calculated using a pre-set algorithm, i.e., the pre-anchor coordinates corresponding to each operator. Finally, the weights corresponding to each operator are used. By fusing all the anchor point coordinates, the anchor point coordinates of the target feature point in the depth image are obtained, i.e.

[0079] It is understood that the preset algorithm in this embodiment also includes a variety of operators, which will not be described in detail here.

[0080] It is also understood that, since this embodiment is for the processing of depth images, the operators used are different from those in the RGB images described above.

[0081] In practice, RGB and depth images are typically processed synchronously during distance measurement or target detection to simultaneously calculate the coordinates of anchor points in each image. Specifically, for example... Figure 7 As shown.

[0082] Figure 7 An anchor point positioning method according to some embodiments of this application may include:

[0083] S01, acquire RGB image and depth image.

[0084] S02, based on dual-camera calibration, obtains the mapping relationship of feature points in RGB and depth images.

[0085] S03, obtain the pixel coordinates of the target feature points in the RGB image, as well as the detection area.

[0086] S04. Based on the mapping relationship, the pixel coordinates and detection area of ​​the target feature points in the depth image are obtained.

[0087] S05, calculating anchor point coordinates of the target feature point in the RGB image and the depth image respectively based on a preset algorithm.

[0088] Specifically, for the obtained RGB image and depth image, the mapping relationship of the feature points of the RGB image and the depth image is obtained according to the dual-camera calibration, and the pixel coordinates of the target feature point indicated by the user are recognized, and then the size of the detection region of the target feature point in the RGB image is obtained according to the user's pre-setting or the user's current input. Further, according to the mapping relationship of the feature points of the RGB image and the depth image, the pixel coordinates of the target feature point in the depth image and the detection region are obtained. Finally, the anchor point corresponding to the target feature point in the detection region is calculated by using a preset algorithm, that is, a calculation module including multiple operators, to obtain the anchor point coordinate values of the target feature point in the RGB image and the depth image.

[0089] It can be understood that each of the above steps is described in detail in the above embodiment, which will not be repeated here.

[0090] The embodiment of the present application sets the detection region corresponding to the target feature point in the RGB image and the depth image, and then calculates the accurate anchor point position corresponding to the target feature point in the detection region in the RGB image and the depth image by using a preset algorithm, to realize accurate positioning of the anchor point, and provide a reliable basis for subsequent ranging, target positioning and other applications.

[0091] Figure 8 The structure diagram of the anchor point positioning device of the embodiment of the present application is shown in the figure, and the device 400 includes:

[0092] The first acquisition module 410 is configured to acquire the pixel coordinates of the target feature point in the RGB image.

[0093] The determination module 420 is configured to determine the detection region corresponding to the target feature point in the RGB image based on the pixel coordinates.

[0094] The first calculation module 430 is configured to calculate the anchor point coordinates corresponding to the target feature point in the detection region in the RGB image based on a preset algorithm.

[0095] Optionally, the anchor point positioning device provided by the embodiment of the present application further includes:

[0096] The second acquisition module 440 is configured to acquire the pixel coordinates of the target feature point in the depth image and the detection region corresponding to the target feature point in the depth image.

[0097] The second calculation module 450 is configured to calculate the anchor point coordinates of the target feature point in the detection region in the depth image based on a preset algorithm.

[0098] Optionally, the anchor point positioning apparatus provided by the embodiment of the present application includes a plurality of operators in the preset algorithm, and the first calculation module or the second calculation module is specifically configured to:

[0099] calculate the pre-anchor point coordinates corresponding to the target feature point based on each operator;

[0100] fuse all the obtained pre-anchor point coordinates based on the weight corresponding to each operator to obtain the anchor point coordinates corresponding to the target feature point.

[0101] Optionally, the anchor point positioning apparatus provided by the embodiment of the present application is obtained through machine learning training.

[0102] Optionally, the anchor point positioning apparatus provided by the embodiment of the present application is obtained through machine learning training.

[0103] obtain the mapping relationship between the RGB image and the depth image based on double-camera calibration;

[0104] obtain the pixel coordinates of the target feature point in the depth image and the corresponding detection region of the target feature point in the depth image based on the pixel coordinates of the target feature point and the mapping relationship.

[0105] Optionally, the anchor point positioning apparatus provided by the embodiment of the present application is obtained through machine learning training.

[0106] set the size of the detection region based on the experience value;

[0107] obtain the detection region based on the size and the pixel coordinates.

[0108] On the other hand, the embodiment of the present application also provides an electronic device, which includes a first coil, a second coil, a near field communication module, a wireless charging receiving module, and a processing device. The processing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is configured to execute the program to implement the anchor point positioning method described in the above embodiment.

[0109] The following refers to Figure 9 , Figure 9 is a structural schematic diagram of the computer electronic device of the processing device of the embodiment of the present application.

[0110] As Figure 9As shown, the computer electronic device 500 includes a central processing unit (CPU) 501 which can perform various appropriate actions and processes in accordance with programs stored in a read only memory (ROM) 502 or programs loaded from the storage section 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) port 505 is also connected to the bus 504.

[0111] Connected to the I / O port 505 are an input section 506 including a keyboard, a mouse, and the like; an output section 507 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage section 508 including a hard disk, and the like; and a communication section 509 including a network port card such as a LAN card, a modem, and the like. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O port 505 as necessary. A removable media 511 such as a magnetic disk, a magneto-optical disk, a semiconductor memory, and the like is attached to the drive 510 as necessary, so that a computer program read therefrom is installed into the storage section 508 as necessary.

[0112] In particular, in accordance with embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a machine-readable medium, the computer program comprising program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 509, and / or installed from the removable media 511. When the computer program is executed by the central processing unit (CPU) 501, the above-described functions defined in the electronic device of the present application are performed.

[0113] It should be noted that the computer-readable medium shown in the application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor electronic device, device or apparatus, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution electronic device, device or apparatus. In this application, the computer-readable signal medium can include a data signal carried in a baseband or as a carrier wave part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate or transmit programs for use by or in conjunction with an instruction execution electronic device, device or apparatus. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0114] The flowcharts and block diagrams in the drawings illustrate the possible implementation architecture, function and operation of the processing device, method and computer program product according to various embodiments of the application. In this regard, each block in the flowchart or block diagram can represent a module, program segment, or a portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order than that shown in the figure. For example, two blocks that are shown in succession can actually be executed substantially concurrently, or they can sometimes be executed in reverse order, depending on the involved functions. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based electronic device that performs the specified function or operation, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0115] The units or modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. The described units or modules can also be arranged in a processor, for example, can be described as: a processor comprising an acquisition module, a determination module and a calculation module. In some cases, the names of these units or modules do not constitute a limitation on the units or modules themselves, for example, the calculation module can also be described as "for calculating the anchor point coordinates corresponding to the target feature point in the detection region in the RGB image based on a preset algorithm".

[0116] As another aspect, the present application also provides a computer readable storage medium, which can be included in the electronic device described in the above embodiments, or can exist independently without being assembled into the electronic device. The computer readable storage medium stores one or more programs, and when the programs are used by one or more processors to execute the anchor point positioning method described in the present application:

[0117] Acquiring pixel coordinates of a target feature point in an RGB image;

[0118] Determining a detection region of the target feature point in the RGB image based on the pixel coordinates;

[0119] Calculating anchor point coordinates corresponding to the target feature point in the detection region in the RGB image based on a preset algorithm.

[0120] In summary, the anchor point positioning method, device, equipment and medium provided by the embodiments of the present application realize the reuse of the coil inside the electronic device by configuring a coil and a switch in the electronic device, then switching the coil connection by the switch according to the current wireless charging state, and switching the high-frequency and low-frequency charging according to the wireless charging equipment during the wireless charging process, which avoids the coil redundancy and improves the charging efficiency.

[0121] The above description is only the preferred embodiments of the present application and the explanation of the applied technical principles. Those skilled in the art should understand that the disclosed range in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and also covers other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the disclosed concept. For example, the above features can be replaced with the technical features disclosed in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. An anchor point positioning method, characterized in that, The method includes: Obtain the pixel coordinates of the target feature points in the RGB image; Based on the pixel coordinates, the detection region corresponding to the target feature point is determined in the RGB image; Based on a preset algorithm, the anchor point coordinates corresponding to the target feature point are calculated within the detection area of ​​the RGB image. Acquire a depth image, and obtain the mapping relationship between the feature points of the RGB image and the depth image based on dual-camera calibration; Based on the mapping relationship, the pixel coordinates and detection area of ​​the target feature point in the depth image are obtained; The anchor point coordinates of the target feature point in the depth image are calculated based on the preset algorithm; The preset algorithm includes multiple operators, which calculate the anchor coordinates corresponding to the target feature point within the detection region of the RGB image, or calculate the anchor coordinates of the target feature point in the depth image, based on the preset algorithm, including: Calculate the coordinates of the pre-anchor point corresponding to the target feature point based on each operator; Based on the weights corresponding to each operator, the coordinates of all the pre-anchor points are fused to obtain the anchor point coordinates corresponding to the target feature point.

2. The anchor point positioning method according to claim 1, characterized in that, The weights are obtained through machine learning training.

3. The anchor point positioning method according to any one of claims 1-2, characterized in that, Determining the detection region corresponding to the target feature point includes: The size of the detection area is set based on empirical values; The detection area is obtained based on the stated size and pixel coordinates.

4. An anchor point positioning device, characterized in that, The device includes: The acquisition module is used to acquire the pixel coordinates of the target feature points in the RGB image and the depth image, and to obtain the mapping relationship between the feature points of the RGB image and the depth image based on dual-camera calibration; The determination module is used to determine the detection region of the target feature point in the RGB image based on the pixel coordinates, and to obtain the pixel coordinates and detection region of the target feature point in the depth image based on the mapping relationship; The calculation module is used to calculate the anchor point coordinates corresponding to the target feature point within the detection area of ​​the RGB image based on a preset algorithm; and to calculate the anchor point coordinates of the target feature point in the depth image based on the preset algorithm. The preset algorithm includes multiple operators, which calculate the anchor coordinates corresponding to the target feature point within the detection region of the RGB image, or calculate the anchor coordinates of the target feature point in the depth image, based on the preset algorithm, including: Calculate the coordinates of the pre-anchor point corresponding to the target feature point based on each operator; Based on the weights corresponding to each operator, the coordinates of all the pre-anchor points are fused to obtain the anchor point coordinates corresponding to the target feature point.

5. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the anchor point positioning method as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon for implementing the anchor point positioning method as described in any one of claims 1-3.

Images