Pan / tilt calibration method and device for valve hall equipment inspection robot, and storage medium

Abstract

A method and device for calibrating a pan-tilt platform of a valve hall equipment inspection robot, and a storage medium, the method comprising: when the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, obtaining a target image captured by the target inspection robot through a camera; identifying the valve hall equipment in the target image to obtain the identified target equipment; determining the image coordinates of the target equipment in the target image; obtaining the preset coordinates corresponding to the target equipment from a database, and calculating the coordinate offset based on the preset coordinates and the above-mentioned image coordinates; calculating the position adjustment amount and the posture adjustment amount for the above-mentioned pan-tilt platform based on the coordinate offset; adjusting the above-mentioned preset position according to the position adjustment amount, and / or adjusting the above-mentioned preset posture according to the posture adjustment amount. By implementing the embodiments of the present application, the position and posture of the robot pan-tilt platform can be calibrated, eliminating the adverse effects of the accumulated offset on the shooting of the target equipment.

Description

Technical Field

[0001] The present application relates to the technical field of electric power equipment, and in particular to a pan / tilt calibration method and device, and a storage medium for a valve hall equipment inspection robot. Background Art

[0002] As an important part of the national power grid, the ultra-high voltage direct current transmission system plays a strategic role in allocating power resources across regions, and the inspection and maintenance of valve hall equipment is the core task to ensure the normal operation of the ultra-high voltage direct current transmission system. In practice, it is found that when using a valve hall equipment inspection robot to perform automated inspections on valve hall equipment, the pan / tilt carrying the robot will accumulate mechanical offsets as the running time increases, which will cause the robot to easily lose the target equipment in its field of view when inspecting at the preset position. Summary of the invention

[0003] The embodiments of the present application disclose a pan-tilt calibration method and device, and a storage medium for a valve hall equipment inspection robot, which can calibrate the position and posture of the robot pan-tilt, eliminating the adverse effects of accumulated offset on the target equipment being photographed.

[0004] The first aspect of the embodiment of the present application discloses a pan / tilt calibration method of a valve hall equipment inspection robot, comprising:

[0005] When the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, a target image captured by the camera of the target inspection robot is obtained;

[0006] Identify the valve hall device in the target image to obtain the identified target device;

[0007] determining image coordinates of the target device in the target image;

[0008] Acquire preset coordinates corresponding to the target device from a database, and calculate a coordinate offset according to the preset coordinates and the image coordinates;

[0009] Calculating a position adjustment amount and a posture adjustment amount for the gimbal according to the coordinate offset;

[0010] The preset position is adjusted according to the position adjustment amount, and / or the preset posture is adjusted according to the posture adjustment amount.

[0011] As an optional implementation, in the first aspect of the embodiment of the present application, adjusting the preset position according to the position adjustment amount, or adjusting the preset posture according to the posture adjustment amount, includes:

[0012] Determining whether the posture adjustment amount is greater than a preset adjustment threshold;

[0013] When the amount is greater than the preset adjustment threshold, a first moving direction and a first moving distance corresponding to the position adjustment amount are obtained, and the gimbal is controlled to move the first moving distance toward the first moving direction, and the preset position is updated according to the position of the gimbal after the movement;

[0014] When it is not greater than the preset adjustment threshold, the first rotation direction and the first rotation angle corresponding to the posture adjustment amount are obtained, and the pan-tilt head is controlled to rotate so that the camera rotates the first rotation angle toward the first rotation direction, and the preset posture is updated according to the posture of the pan-tilt head after rotation.

[0015] As another optional implementation, in the first aspect of the embodiment of the present application, adjusting the preset position according to the position adjustment amount, and adjusting the preset posture according to the posture adjustment amount, includes:

[0016] Acquire a second moving direction and a second moving distance corresponding to the position adjustment amount, and control the pan / tilt head to move the second moving distance toward the second moving direction;

[0017] Acquire a second rotation direction and a second rotation angle corresponding to the posture adjustment amount, and control the pan / tilt head to rotate so that the camera rotates toward the second rotation direction and the second rotation angle;

[0018] The preset position is updated according to the position of the pan-tilt platform after movement, and the preset posture is updated according to the posture of the pan-tilt platform after rotation.

[0019] As another optional implementation, in the first aspect of the embodiment of the present application, the adjusting the preset position according to the position adjustment amount, and / or adjusting the preset posture according to the posture adjustment amount includes:

[0020] When the coordinate offset is greater than a preset deviation threshold, the preset position is adjusted according to the position adjustment amount, and / or the preset posture is adjusted according to the posture adjustment amount, and the process of obtaining the target image captured by the target inspection robot through the camera is continued until the coordinate offset is no greater than the preset deviation threshold.

[0021] As another optional implementation, in the first aspect of the embodiment of the present application, after adjusting the preset position according to the position adjustment amount, and / or adjusting the preset posture according to the posture adjustment amount, and continuing to obtain the target image captured by the target inspection robot through the camera until the coordinate offset is not greater than the preset deviation threshold, the method further includes:

[0022] Acquire multiple groups of adjustment data calculated in the process of adjusting the preset position and the preset posture, each group of adjustment data includes the coordinate offset, the position adjustment and the posture adjustment calculated each time;

[0023] When the number of the plurality of groups of adjustment data is greater than a preset group number threshold, mapping relationships between the coordinate offset and the position adjustment and the posture adjustment are fitted according to the plurality of groups of adjustment data;

[0024] The mapping relationship is stored in a database.

[0025] As another optional implementation, in the first aspect of the embodiment of the present application, identifying the valve hall device in the target image to obtain the identified target device includes:

[0026] According to the valve hall equipment characteristic data stored in the database, identifying one or more valve hall equipment in the target image that matches the valve hall equipment characteristic data;

[0027] The distances between the one or more valve hall devices and the center of the target image in the target image are calculated, and a valve hall device that is closest to the center of the target image is used as the identified target device.

[0028] As another optional implementation, in the first aspect of the embodiment of the present application, when it is identified that there is no valve hall device in the target image according to the valve hall device feature data stored in the database, the method further includes:

[0029] Identify whether there are one or more suspected image blocks in the target image, wherein feature data of the suspected image blocks partially matches feature data of the valve hall device;

[0030] Acquire one with the highest matching degree from the one or more suspected image blocks as the target image block;

[0031] According to the position of the target image block in the target image, the pan / tilt is controlled to drive the camera to rotate, and the step of obtaining the target image captured by the target inspection robot through the camera is continued until at least one valve hall device is identified.

[0032] A second aspect of an embodiment of the present application discloses a pan / tilt calibration device for a valve hall equipment inspection robot, comprising:

[0033] An acquisition unit, used for acquiring a target image captured by a camera of the target inspection robot when the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture;

[0034] An identification unit, used for identifying the valve hall device in the target image to obtain the identified target device;

[0035] a determining unit, configured to determine the image coordinates of the target device in the target image;

[0036] A first calculation unit, configured to obtain preset coordinates corresponding to the target device from a database, and calculate a coordinate offset according to the preset coordinates and the image coordinates;

[0037] A second calculation unit, used for calculating a position adjustment amount and a posture adjustment amount for the gimbal according to the coordinate offset;

[0038] An adjustment unit is used to adjust the preset position according to the position adjustment amount, and / or to adjust the preset posture according to the posture adjustment amount.

[0039] A third aspect of the present application discloses an electronic device, including:

[0040] A memory storing executable program code;

[0041] a processor coupled to the memory;

[0042] The processor calls the executable program code stored in the memory to execute all or part of the steps in any one of the pan / tilt calibration methods for the valve hall equipment inspection robot disclosed in the first aspect of the embodiment of the present application.

[0043] The fourth aspect of an embodiment of the present application discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute all or part of the steps in any pan-tilt calibration method for a valve hall equipment inspection robot disclosed in the first aspect of an embodiment of the present application.

[0044] A fifth aspect of an embodiment of the present application discloses a computer program product. When the computer program product runs on a computer, the computer executes all or part of the steps in any one of the pan-tilt calibration methods for a valve hall equipment inspection robot in the first aspect of the embodiment of the present application.

[0045] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0046] In the embodiment of the present application, the image captured by the valve hall equipment inspection robot can be identified by image recognition, and the change of the equipment position can be compared to obtain the robot's pan-tilt offset, and then the pan-tilt can be calibrated according to the offset. It can be seen that by implementing the embodiment of the present application, the position and posture of the robot pan-tilt can be calibrated to eliminate the adverse effects of the accumulated offset on the target equipment being photographed. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 It is a schematic diagram of an application scenario of a pan / tilt calibration method of a valve hall equipment inspection robot disclosed in an embodiment of the present application;

[0049] Figure 2 It is a flow chart of a pan / tilt calibration method of a valve hall equipment inspection robot disclosed in an embodiment of the present application;

[0050] Figure 3 is a schematic diagram of a coordinate offset disclosed in an embodiment of the present application;

[0051] Figure 4 It is a flow chart of another pan / tilt calibration method of a valve hall equipment inspection robot disclosed in an embodiment of the present application;

[0052] Figure 5 It is a flowchart of another pan / tilt calibration method of a valve hall equipment inspection robot disclosed in an embodiment of the present application;

[0053] Figure 6 It is a modular schematic diagram of a pan-tilt calibration device of a valve hall equipment inspection robot disclosed in an embodiment of the present application;

[0054] Figure 7 It is a modular schematic diagram of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] It should be noted that the terms "including" and "having" in the embodiments of the present application and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0057] The embodiment of the present application discloses a pan-tilt calibration method and device and storage medium for a valve hall equipment inspection robot, which can calibrate the position and posture of the robot pan-tilt to eliminate the adverse effects of accumulated offset on the target equipment being photographed. Detailed description will be given below in conjunction with the accompanying drawings.

[0058] See also Figure 1 , Figure 1 It is a schematic diagram of an application scenario of a pan-tilt calibration method for a valve hall equipment inspection robot disclosed in an embodiment of the present application, including a host computer 101, a track 102, a pan-tilt 103, and an inspection robot 104. Among them, the host computer 101 can establish a communication connection with the pan-tilt 103 to control the movement and rotation of the pan-tilt 103; the pan-tilt 103 is set on the track 102, and can move along the laying direction of the track 102; the pan-tilt 103 is also connected to the inspection robot 104, for carrying the inspection robot 104, and under the control of the host computer 101, can drive the inspection robot 104 to move along the track, or drive the inspection robot 104 to rotate so that the camera it carries is facing the valve hall equipment to be photographed, for shooting and detection. It can be understood that Figure 1 The track 102 shown is a vertical track, which is only an example. In some embodiments, the track 102 may also be a horizontal track, an inclined track, etc.

[0059] Optionally, a track slide 105 may also be provided on the track 102, and the pan-tilt head 103 may be provided on the track slide 105, so that the pan-tilt head 103 moves along the laying direction of the track 102. In fact, the track slide 105 may slide relative to the track 102 and synchronously drive the pan-tilt head 103 to move relative to the track 102.

[0060] In this embodiment, the host computer 101 can be pre-set at one or more preset positions on the track 102, and a preset posture corresponding to each preset position. When the pan-tilt 103 reaches a preset position under the communication control of the host computer 101, it can be adjusted to a preset posture corresponding to the preset position, so that the inspection robot 104 connected to the pan-tilt 103 can stay at the preset position, and the camera carried by it can face the direction corresponding to the preset posture to shoot and detect the valve hall equipment at the direction. On this basis, the host computer 101 can obtain the image captured by the above-mentioned camera, and adjust the preset position and preset posture of the above-mentioned pan-tilt 103 according to the image position of the valve hall equipment displayed on the image, so as to realize the calibration of the position and posture of the pan-tilt 103, and eliminate the adverse effects of the accumulated offset generated by the pan-tilt 103 after long-term operation on the shooting of the valve hall equipment (such as the image position offset of the valve hall equipment on the captured image, image incompleteness, etc.).

[0061] Figure 2 A schematic flow chart of a pan / tilt calibration method for a valve hall equipment inspection robot is disclosed, and the method can be applied to the above-mentioned host computer. Figure 2 As shown, the method may include the following steps:

[0062] 201. When the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, a target image captured by the target inspection robot through a camera is obtained.

[0063] In the embodiment of the present application, the valve hall refers to a building used to arrange the converter valve and its related equipment in the DC power transmission system, and the valve hall equipment may include the converter valve, voltage divider, mutual inductor, etc. The positions of various valve hall equipment in the valve hall are usually fixed, and the relative position relationship between different valve hall equipment is usually fixed. Therefore, the valve hall equipment can be inspected by an inspection robot that automatically runs on the set inspection route, so that the working status of the valve hall equipment can be monitored in real time and abnormalities can be discovered in time. It can be understood that a track for the operation of the pan-tilt can be laid in advance on the above-set inspection route, so that the above-mentioned inspection robot can move along the track on the inspection route driven by the pan-tilt connected thereto. Among them, the above-mentioned pan-tilt is used to carry the inspection robot, drive the inspection robot to move, and control the rotation of the inspection robot; the above-mentioned inspection robot can carry a camera (such as a visible light camera, an infrared camera, etc.) to shoot and detect the operating status of the above-mentioned valve hall equipment according to the control instructions of the host computer.

[0064] Among them, one or more preset positions for the pan-tilt may be provided on the above-mentioned track. For example, for valve hall equipment in a certain area, when the pan-tilt connected to the inspection robot runs on the track set in the area, and the inspection robot can photograph any valve hall equipment, the host computer can obtain the track segment where the pan-tilt is located at this time, and use the midpoint of each track segment in the one or more obtained track segments as the preset position. In some embodiments, the endpoints, trisection points, etc. of each track segment can also be obtained as the above-mentioned preset positions. For example, when there are two valve hall devices in a certain area, at least two track segments can be divided from the track set in the area, wherein the inspection robot can photograph at least one valve hall device on each track segment; optionally, when there is more than one track segment corresponding to a certain valve hall device according to the above division method, a track segment with the best shooting effect for the valve hall device (such as the highest clarity of the captured image, the highest degree of centering of the valve hall device in the captured image, etc.) can be selected from the more than one track segments, and other track segments can be eliminated at the same time; finally, the midpoint of each of the remaining at least two track segments can be used as a preset position, so that each valve hall device in the above area can correspond to at least one preset position, and when the pan / tilt is in the preset position, the inspection robot can photograph the corresponding valve hall device. It can be understood that each of the above track segments can be independent of each other or partially overlap.

[0065] Furthermore, the gimbal may correspond to one or more preset postures at each preset position, wherein the preset postures may include the azimuth angle (i.e., the angle of the gimbal's lateral rotation) and the pitch angle (i.e., the angle of the gimbal's vertical rotation). Exemplarily, when the gimbal moves to a preset position, the valve hall device corresponding to the preset position may be photographed in different postures (i.e., different azimuth angles and pitch angles) to obtain multiple photographed images, and one or more photographed images whose photographic effects meet preset requirements (such as the clarity of the photographed image is higher than a preset threshold, the distance between the position of the valve hall device in the photographed image and the center of the image is lower than another preset threshold, etc.) are selected, and the corresponding postures are used as one or more preset postures corresponding to the preset position. For example, when a valve hall device occupies a larger space (such as a large-scale flow-converting valve group), the pan-tilt head reaches the preset position corresponding to the valve hall device, and then can be adjusted to a plurality of different postures in turn so that the inspection robot can photograph the valve hall device from different angles; by analyzing the multiple images taken, a clear image with less noise and in which the above-mentioned valve hall device is located near the center of the image can be selected, and the posture of the pan-tilt head when taking the image can be used as the preset posture.

[0066] Furthermore, the preset position and preset posture can be set at the same time, that is, by continuously adjusting the position of the pan-tilt on the track and the posture of the pan-tilt itself, the position and posture of the pan-tilt when the valve hall equipment has the best shooting effect can be set as the preset position and preset posture respectively. For example, when the pan-tilt is running on the track, its posture can be continuously adjusted so that a certain valve hall equipment always exists in the shooting preview screen of the inspection robot, and the position, size, clarity, etc. of the valve hall equipment in the shooting preview screen are scored in real time, and the position and posture of the pan-tilt when the comprehensive score is the highest are recorded as the preset position and preset posture respectively. Specifically, the distance between the position of the valve hall equipment in the shooting preview screen and the center of the screen can be recorded as Score1, and the weight coefficient can be preset as a; the area ratio of the valve hall equipment in the shooting preview screen can be recorded as Score2, and the weight coefficient can be preset as b; the magnification of the shooting preview screen can be recorded as Score3, and the weight coefficient can be preset as c. Among them, the weight coefficient can be preset according to the importance of the corresponding factor in the scoring. On this basis, the comprehensive score S = Score1·a + Score3·b + Score3·c can be obtained according to the weighted sum formula, and according to the change of the comprehensive score S when the pan-tilt head is adjusted, the position and posture of the above-mentioned pan-tilt head when the comprehensive score S is the highest can be updated in real time to obtain the preset position and preset posture corresponding to the above-mentioned valve hall equipment after the pan-tilt head runs once.

[0067] On this basis, when the pan / tilt connected to the target inspection robot reaches a preset position, it can be adjusted to a preset posture corresponding to the preset position, and the camera of the target inspection robot can be mobilized to shoot and obtain the target image. The type of the above-mentioned target image is determined by the type of camera, which can be a visible light image, an infrared image, or a visible light-infrared fusion image, which is not specifically limited in the embodiments of the present application. Generally speaking, the above-mentioned target image contains at least one valve hall device.

[0068] As an optional implementation, when the pan-tilt head connected to the target inspection robot reaches a preset position, the posture of the pan-tilt head can also be manually adjusted according to the manual control instructions issued by the host computer, so that the target inspection robot can shoot in the direction that the user needs to shoot, which is conducive to improving the flexibility of the robot inspection.

[0069] 202. Identify the valve hall device in the target image to obtain the identified target device.

[0070] Specifically, the host computer may store image feature data of various valve hall equipment in a database in advance. The image feature data may be image feature data corresponding to the image of the valve hall equipment taken by the inspection robot connected to the pan-tilt when the pan-tilt is in the above-mentioned preset position and in the above-mentioned preset posture. In some embodiments, the host computer may obtain the image feature data corresponding to the target image, and compare it with the image feature data of various valve hall equipment, and use the valve hall equipment with the highest matching rate as the target equipment identified from the target image. In other embodiments, the host computer may directly subtract the target image from the images of various valve hall equipment, and perform grayscale statistics on the difference image obtained by the subtraction, and use the valve hall equipment corresponding to the difference image with the lowest grayscale statistical value as the target equipment.

[0071] 203. Determine the image coordinates of the target device in the target image.

[0072] Specifically, the image coordinates may include a horizontal coordinate and a vertical coordinate, which are used to represent the position information of the target device in the target image. The image coordinates may be the coordinates of a point of the target device on the target image, such as a center point, a vertex, etc. Figure 3 As shown on the left, in a rectangular coordinate system established with the lower left vertex of the target image as the origin, the coordinates of a vertex of the target device on the target image are (x1, y1), and the coordinates can be used as the image coordinates of the target device on the target image.

[0073] 204. Obtain preset coordinates corresponding to the target device from a database, and calculate a coordinate offset according to the preset coordinates and the image coordinates.

[0074] See also Figure 3 , Figure 3 is a schematic diagram of a coordinate offset disclosed in this embodiment. Figure 3 As shown on the left, by executing the above step 203, the coordinates (x1, y1) of a vertex of the target device on the target image (i.e., a position on the entity corresponding to the target device) can be used as the image coordinates of the target device; Figure 3 As shown on the right, the preset coordinates corresponding to the target device refer to the coordinates (x2, y2) of the above-mentioned vertex of the target device on the preset image, wherein the above-mentioned preset image may be an image obtained by the inspection robot taking a picture of the target device through a camera when the pan / tilt is at the above-mentioned preset position for the first time and at the above-mentioned preset posture. It can be understood that the above-mentioned image coordinates and the preset coordinates correspond to the same vertex of the target device on the target image and the preset image respectively, which is only an example, and the image coordinates and the preset coordinates may also correspond to the center point, identification point, etc. of the target device, as long as they correspond to the same position on the entity of the target device.

[0075] According to the above image coordinates (x1, y1) and preset coordinates (x2, y2), the coordinate offset of the target device on the target image can be obtained by difference, including the horizontal axis offset Δx (where Δx = x1-x2) and the vertical axis offset Δy (where Δy = y1-y2). It can be understood that when Δx is greater than 0, it means that the target device has shifted to the right on the target image, and when Δx is less than 0, it means that it has shifted to the left; similarly, when Δy is greater than 0, it means that the target device has shifted upward on the target image, and when Δy is less than 0, it means that it has shifted downward. By calculating the above coordinate offsets, the offset direction and degree of the gimbal when shooting the target device can be determined, thereby providing an accurate data basis for the subsequent calibration steps.

[0076] 205. Calculate the position adjustment amount and attitude adjustment amount for the gimbal according to the coordinate offset.

[0077] Specifically, the above coordinate offset can form a certain functional mapping relationship with the position adjustment amount and attitude adjustment amount of the gimbal, that is, according to the above coordinate offset, a unique position adjustment amount and attitude adjustment amount can be obtained. For example, the above coordinate offset can be in pixels. When the horizontal axis offset Δx is A pixels and the vertical axis offset Δy is B pixels, the position adjustment amount of the gimbal (such as when only vertical adjustment is performed) can be calculated according to the function f 纵向 (A, B) is determined, and the attitude adjustment amount (including azimuth adjustment amount and pitch angle adjustment amount) can be determined according to the function group g 方向 (A,B),g 俯仰 (A, B) are jointly determined.

[0078] As an optional implementation, after calculating the above-mentioned position adjustment amount and attitude adjustment amount for the gimbal, the historical adjustment data of the gimbal can also be obtained from the database of the host computer. The historical adjustment data may include the position adjustment amount record, attitude adjustment amount record, adjustment times, etc. of the gimbal, and then the above-mentioned position adjustment amount and attitude adjustment amount can be corrected according to the historical adjustment data. For example, when the attitude adjustment amount record of the gimbal indicates that the gimbal has performed pitch angle adjustment many times, and its pitch angle adjustment amount fluctuates around a fixed value, and the number of adjustments is more than a preset threshold, it can be determined that the gimbal has a fixed control offset on the pitch angle, so that the attitude adjustment amount of the gimbal can be corrected according to the above-mentioned fixed value. By executing the above method, the gimbal calibration can be performed with reference to historical data, thereby improving the accuracy of the gimbal calibration.

[0079] Optionally, the host computer may also generate fault information including the above-mentioned fixed control offset, and send the fault information to maintenance personnel so that the maintenance personnel can determine whether manual maintenance is required.

[0080] 206. Adjust the preset position according to the position adjustment amount, and / or adjust the preset posture according to the posture adjustment amount.

[0081] Specifically, the host computer can generate a first adjustment instruction according to the above-mentioned position adjustment amount, so as to control the servo driver to generate a corresponding motor drive signal according to the first adjustment instruction, and drive the servo motor to drive the pan-tilt to move the distance corresponding to the above-mentioned position adjustment amount on the track. Optionally, when the above-mentioned pan-tilt is set on a track slide, the above-mentioned servo motor is used to drive the track slide to move, so as to indirectly drive the above-mentioned pan-tilt to move. Similarly, the host computer can generate a second adjustment instruction according to the above-mentioned attitude adjustment amount, so as to control the azimuth drive and pitch drive inside the pan-tilt to generate corresponding motor drive signals respectively according to the second adjustment instruction, drive the motor that controls the pan-tilt azimuth angle adjustment to make the pan-tilt rotate horizontally, and drive the motor that controls the pan-tilt angle pitch angle to make the pan-tilt rotate vertically. On this basis, the above-mentioned preset position can be adjusted (i.e. updated) according to the position of the pan-tilt after movement, and the above-mentioned preset attitude can be adjusted according to the attitude of the pan-tilt after rotation.

[0082] It should be noted that, in the process of executing the above step 206, only the preset position of the gimbal may be adjusted, only the preset posture of the gimbal may be adjusted, or both the preset position and the preset posture of the gimbal may be adjusted at the same time.

[0083] As an optional implementation, during the above adjustment of the pan-tilt head, the shooting preview picture of the inspection robot's camera can also be obtained, and the target device can be judged whether it is in a better shooting position based on the shooting preview picture. If it is in a better shooting position, the adjustment is stopped immediately. Among them, the above-mentioned better shooting position can refer to the center point of the target device being near the center of the shooting preview picture (such as within a circular range with the center of the shooting preview picture as the center and 1 / 3 of the frame as the radius). By executing the above method, when the inspection robot is conducting an emergency inspection, the pan-tilt head calibration time can be shortened, the target device can be quickly captured, and shooting and detection can be performed, which is conducive to improving the flexibility of calibration.

[0084] It can be seen that by implementing the method described in the above embodiment, the images taken by the valve hall equipment inspection robot can be identified by image recognition, and the changes in the equipment positions can be compared, so as to obtain the robot's pan-tilt offset, and then the pan-tilt can be calibrated according to the offset. It can be seen that by implementing the embodiment of the present application, the position and posture of the robot pan-tilt can be calibrated, eliminating the adverse effects of the accumulated offset on the target equipment being photographed.

[0085] See also Figure 4 , Figure 4FIG. 1 is a flow chart of another method for calibrating the pan / tilt platform of a valve hall equipment inspection robot disclosed in an embodiment of the present application, which method can be applied to the above-mentioned host computer. Figure 4 As shown, the method may include the following steps:

[0086] 401. When the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, a target image captured by the target inspection robot through a camera is obtained.

[0087] 402. Identify the valve hall device in the target image to obtain the identified target device.

[0088] 403. Determine the image coordinates of the target device in the target image.

[0089] 404. Obtain preset coordinates corresponding to the target device from a database, and calculate a coordinate offset based on the preset coordinates and the image coordinates.

[0090] 405. Calculate the position adjustment amount and attitude adjustment amount for the gimbal according to the coordinate offset.

[0091] Among them, steps 401-405 are similar to the above steps 201-205 and will not be repeated here.

[0092] 406. Determine whether the posture adjustment amount is greater than a preset adjustment threshold.

[0093] The preset adjustment threshold may be an upper limit of the attitude adjustment amount set to avoid excessive attitude adjustment of the gimbal. When the attitude adjustment amount is greater than the preset adjustment threshold, the position of the gimbal may be adjusted directly, replacing the attitude adjustment with a smaller position adjustment, thereby maintaining the stability of the gimbal as much as possible and reducing the occurrence of potential problems such as image jitter and loss of focus; when the attitude adjustment amount is not greater than the preset adjustment threshold, the gimbal attitude is adjusted directly.

[0094] 407. When the amount is greater than the preset adjustment threshold, obtain a first moving direction and a first moving distance corresponding to the position adjustment amount, control the pan / tilt platform to move the first moving distance toward the first moving direction, and update the preset position according to the position of the pan / tilt platform after movement.

[0095] Specifically, the first moving direction may be determined according to the type of track the PTZ is currently located in. For example, if the track the PTZ is currently located in is a vertical track, the first moving direction may be upward or downward.

[0096] In particular, if the track where the gimbal is currently located is spliced ​​with other tracks in different directions, and the gimbal can move across tracks, then the first moving direction may include a set of directions in which the gimbal moves sequentially on different tracks. In this case, the first moving distance may also include a set of distances in which the gimbal moves sequentially on different tracks, wherein each distance value corresponds one-to-one to each direction in the above direction set.

[0097] 408. When it is not greater than the preset adjustment threshold, obtain the first rotation direction and the first rotation angle corresponding to the posture adjustment amount, and control the rotation of the gimbal to rotate the camera toward the first rotation direction by the first rotation angle, and update the preset posture according to the posture of the gimbal after rotation.

[0098] As an optional implementation, before executing the above steps 406-408, it is possible to first determine whether the above coordinate offset is greater than a preset deviation threshold. If not, it means that the inspection robot can still obtain a good shooting effect when shooting the target device through the camera, so the above steps 406-408 can be omitted to shorten the inspection time and reduce unnecessary gimbal mechanical wear.

[0099] When it is determined that the above-mentioned coordinate offset is greater than the preset deviation threshold, the above-mentioned steps 406-408 can be executed to adjust the position and posture of the gimbal; next, the above-mentioned steps 401-405 can be executed in a loop, that is, the target image captured by the target inspection robot through the camera is re-acquired, and then the coordinate offset is recalculated; when the coordinate offset is still greater than the above-mentioned preset deviation threshold, the above-mentioned steps are repeated until the calculated coordinate offset is no greater than the above-mentioned preset deviation threshold.

[0100] Furthermore, the host computer can also obtain multiple groups of adjustment data calculated in the process of adjusting the above-mentioned preset position and the preset posture, each group of adjustment data includes the coordinate offset, position adjustment and posture adjustment calculated each time; then, when the number of the above-mentioned multiple groups of adjustment data is greater than the preset group number threshold, the mapping relationship between the coordinate offset and the position adjustment and posture adjustment can be fitted according to the above-mentioned multiple groups of adjustment data, and the mapping relationship is stored in the database. On this basis, when the above-mentioned step 405 is executed again, the mapping relationship can be called from the database to calculate the position adjustment and posture adjustment of the gimbal, so as to more accurately calculate the position adjustment and posture adjustment, which is conducive to making the coordinate offset not greater than the above-mentioned preset deviation threshold within a fewer number of adjustments, thereby shortening the gimbal calibration time.

[0101] It can be seen that by implementing the method described in the above embodiment, the position and posture of the gimbal are adjusted back and forth multiple times, so that the offset of the gimbal can be reduced to a certain smaller threshold, thereby realizing accurate calibration of the position and posture of the robot gimbal, and accurately restoring it to the preset position, which is beneficial to eliminating the adverse effects of the accumulated offset on the shooting target device; at the same time, by learning the mapping relationship between the coordinate offset and the position adjustment amount and the posture adjustment amount, respectively, the adjustment amount can be optimized with the help of historical data, thereby reducing the number of adjustments as much as possible and shortening the gimbal calibration time.

[0102] See also Figure 5 , Figure 5 FIG. 1 is a flow chart of another method for calibrating the pan / tilt platform of a valve hall equipment inspection robot disclosed in an embodiment of the present application, which method can be applied to the above-mentioned host computer. Figure 5 As shown, the method may include the following steps:

[0103] 501. When the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, a target image captured by the target inspection robot through a camera is obtained.

[0104] Among them, step 501 is similar to the above step 201 and will not be repeated here.

[0105] 502. According to the valve hall equipment characteristic data stored in the database, identify one or more valve hall equipment in the target image that matches the valve hall equipment characteristic data;

[0106] Wherein, step 502 is similar to the above step 202. It should be noted that the above valve hall equipment characteristic data may include image characteristic data, temperature characteristic data, radiation characteristic data, and the like.

[0107] 503. Calculate the distance between the one or more valve hall devices and the center of the target image in the target image, and use the valve hall device closest to the center of the target image as the identified target device.

[0108] By executing the above step 503, the valve hall device closest to the center of the captured target image can be used as the target device.

[0109] As an optional implementation, when it is identified that there is no valve hall device in the above-mentioned target image, it is also possible to identify whether there are one or more suspected image blocks in the target image, and the feature data of the suspected image blocks partially matches the feature data of the above-mentioned valve hall device, which means that there is an incomplete valve hall device in the target image; then, the one with the highest matching degree can be obtained from the above-mentioned one or more suspected image blocks as the target image block; next, according to the position of the target image block in the above-mentioned target image, the pan-tilt head can be controlled to drive the camera of the inspection robot to rotate, and the above-mentioned steps 501-503 can be executed in a loop, that is, the target image taken by the target inspection robot through the camera is re-acquired; when the valve hall device still cannot be identified from the new target image, the above-mentioned steps are repeated until at least one valve hall device is identified.

[0110] Exemplarily, when it is identified that there is no valve hall equipment in the target image acquired for the first time, but there are several image blocks that match the incomplete valve hall equipment, the image block with the highest matching degree can be used as the target image block. The above-mentioned image blocks can be divided into rectangles of equal proportion to the target image, or divided according to the color characteristics of the target image. On this basis, if the target image block is located in the lower left corner of the target image, the pan-tilt head can be controlled to drive the camera of the inspection robot to rotate to the lower left to capture the complete valve hall equipment; similarly, if the target image block is located on the right side of the target image, the pan-tilt head can be controlled to drive the camera of the inspection robot to rotate to the right.

[0111] By executing the above method, when the gimbal is offset greatly and the complete valve hall equipment cannot be captured, the gimbal can be preliminarily calibrated according to some characteristics of the valve hall equipment, and a more detailed calibration can be performed after the complete valve hall equipment is obtained, thereby improving the fault tolerance of the gimbal calibration and making the gimbal calibration more flexible and reliable.

[0112] 504. Determine the image coordinates of the target device in the target image.

[0113] 505. Obtain preset coordinates corresponding to the target device from a database, and calculate a coordinate offset based on the preset coordinates and the above-mentioned image coordinates.

[0114] 506. Calculate the position adjustment amount and attitude adjustment amount for the gimbal according to the coordinate offset.

[0115] Among them, steps 504-506 are similar to the above steps 203-205 and will not be repeated here.

[0116] 507 . Obtain a second moving direction and a second moving distance corresponding to the position adjustment amount, and control the pan / tilt platform to move the second moving distance toward the second moving direction.

[0117] 508. Obtain a second rotation direction and a second rotation angle corresponding to the posture adjustment amount, and control the pan / tilt head to rotate so that the camera rotates toward the second rotation direction by a second rotation angle.

[0118] 509. Update the preset position according to the position of the gimbal after movement, and update the preset posture according to the posture of the gimbal after rotation.

[0119] By executing the above steps 507-509, the position of the gimbal can be adjusted first to achieve a relatively large gimbal calibration; on this basis, the posture of the gimbal (including azimuth and pitch angle) can be adjusted to achieve a relatively small and relatively fine gimbal calibration. By combining the above two adjustment methods, most of the gimbal offsets can be calibrated to improve the shooting effect of the inspection robot shooting the target equipment.

[0120] As an optional implementation, before executing the above steps 507-509, it is possible to first determine whether the above coordinate offset is greater than a preset deviation threshold. If not, it means that the inspection robot can still obtain a good shooting effect when shooting the target device through the camera, so the above steps 507-509 can be omitted to shorten the inspection time and reduce unnecessary gimbal mechanical wear.

[0121] When it is determined that the above-mentioned coordinate offset is greater than the preset deviation threshold, the above-mentioned steps 507-509 can be executed to adjust the position and posture of the gimbal; next, the above-mentioned steps 501-506 can be executed in a loop, that is, the target image captured by the target inspection robot through the camera is re-acquired, and then the coordinate offset is recalculated; when the coordinate offset is still greater than the above-mentioned preset deviation threshold, the above-mentioned steps are repeated until the calculated coordinate offset is no greater than the above-mentioned preset deviation threshold.

[0122] It can be seen that by implementing the method described in the above embodiment, the position and posture of the gimbal are adjusted back and forth multiple times, so that the offset of the gimbal can be reduced to a smaller threshold, thereby realizing accurate calibration of the position and posture of the robot gimbal, and accurately restoring it to the preset position, which is beneficial to eliminating the adverse effects of the accumulated offset on the shooting target device; at the same time, it can also improve the fault tolerance of the gimbal calibration, making the gimbal calibration more flexible and reliable.

[0123] See also Figure 6 , Figure 6 1 is a modular schematic diagram of a pan-tilt calibration device for a valve hall equipment inspection robot disclosed in an embodiment of the present application. Figure 6As shown, the electronic device may include an acquisition unit 601, an identification unit 602, a determination unit 603, a first calculation unit 604, a second calculation unit 605, and an adjustment unit 606, wherein:

[0124] The acquisition unit 601 is used to acquire a target image captured by a camera of the target inspection robot when the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture;

[0125] An identification unit 602 is used to identify the valve hall device in the target image to obtain the identified target device;

[0126] A determination unit 603 is used to determine the image coordinates of the target device in the target image;

[0127] The first calculation unit 604 is used to obtain preset coordinates corresponding to the target device from the database, and calculate the coordinate offset according to the preset coordinates and the image coordinates;

[0128] A second calculation unit 605 is used to calculate a position adjustment amount and a posture adjustment amount for the above-mentioned gimbal according to the coordinate offset;

[0129] The adjustment unit 606 is used to adjust the preset position according to the position adjustment amount, and / or adjust the preset posture according to the posture adjustment amount.

[0130] It can be seen that the device described in the above embodiment can identify the images taken by the valve hall equipment inspection robot through image recognition, and compare the changes in the equipment position therein, so as to obtain the robot's pan-tilt offset, and then calibrate the pan-tilt according to the offset, so as to calibrate the position and posture of the robot's pan-tilt, which is beneficial to eliminate the adverse effects of the accumulated offset on the shooting target equipment.

[0131] As an optional implementation, the adjustment unit 606 in the above embodiment may include a determination subunit and a first adjustment subunit (not shown), wherein:

[0132] A judging subunit, used to judge whether the above-mentioned posture adjustment amount is greater than a preset adjustment threshold;

[0133] The first adjustment subunit is used to obtain the first moving direction and the first moving distance corresponding to the above-mentioned position adjustment amount when the judging subunit judges that it is greater than the above-mentioned preset adjustment threshold, and control the above-mentioned pan-tilt head to move the first moving distance toward the first moving direction, and update the above-mentioned preset position according to the position of the pan-tilt head after movement; when the judging subunit judges that it is not greater than the above-mentioned preset adjustment threshold, obtain the first rotation direction and the first rotation angle corresponding to the above-mentioned posture adjustment amount, and control the above-mentioned pan-tilt head to rotate so that the above-mentioned camera rotates toward the first rotation direction by the first rotation angle, and update the above-mentioned preset posture according to the posture of the pan-tilt head after rotation.

[0134] As another optional implementation, the adjustment unit 606 in the above embodiment may also include a first acquisition subunit, a second acquisition subunit, and a second adjustment subunit, which are not shown in the figure, wherein:

[0135] A first acquisition subunit is used to acquire a second moving direction and a second moving distance corresponding to the position adjustment amount, and control the pan / tilt head to move the second moving distance toward the second moving direction;

[0136] A second acquisition subunit is used to acquire a second rotation direction and a second rotation angle corresponding to the posture adjustment amount, and control the pan / tilt head to rotate so that the camera rotates toward the second rotation direction and the second rotation angle;

[0137] The second adjustment subunit is used to update the preset position according to the position of the pan / tilt head after movement, and to update the preset posture according to the posture of the pan / tilt head after rotation.

[0138] As another optional implementation, the gimbal calibration device in the above embodiment may further include a data acquisition unit, a fitting unit, and a storage unit, which are not shown in the figure, wherein:

[0139] The above-mentioned adjustment unit 606 is specifically used to adjust the preset position according to the above-mentioned position adjustment amount, and / or adjust the preset posture according to the above-mentioned posture adjustment amount when the above-mentioned coordinate offset is greater than the preset deviation threshold, and trigger the acquisition unit 601 to continue to acquire the target image taken by the target inspection robot through the camera until the above-mentioned coordinate offset is no greater than the preset deviation threshold.

[0140] A data acquisition unit, used to acquire multiple groups of adjustment data calculated in the process of adjusting the preset position and the preset posture, each group of adjustment data includes a coordinate offset, a position adjustment amount and a posture adjustment amount calculated each time;

[0141] A fitting unit, configured to fit mapping relationships between the coordinate offsets and the position adjustment amounts and the posture adjustment amounts, respectively, according to the plurality of adjustment data when the number of the plurality of adjustment data is greater than a preset threshold number of groups;

[0142] The storage unit is used to store the mapping relationship in a database.

[0143] It can be seen that the device described in the above embodiment can optimize the adjustment amount with the help of historical data by learning the mapping relationship between the coordinate offset and the position adjustment amount and the attitude adjustment amount, thereby reducing the number of adjustments as much as possible and shortening the gimbal calibration time.

[0144] As another optional implementation, the identification unit 602 in the above embodiment may include a first identification subunit, a calculation subunit, a second identification subunit, a third acquisition subunit, and a control subunit, which are not shown in the figure, wherein:

[0145] A first identification subunit is used to identify one or more valve hall devices in the target image that match the valve hall device characteristic data according to the valve hall device characteristic data stored in the database;

[0146] a calculation subunit, configured to calculate the distance between the one or more valve hall devices and the center of the target image in the target image, and to take the valve hall device closest to the center of the target image as the identified target device;

[0147] A second recognition subunit is used to recognize whether there are one or more suspected image blocks in the target image when the first recognition subunit recognizes that there is no valve hall device in the target image, and the feature data of the suspected image blocks partially matches the feature data of the valve hall device;

[0148] A third acquisition subunit is used to acquire one with the highest matching degree from the one or more suspected image blocks as a target image block;

[0149] The control subunit is used to control the pan-tilt head to drive the camera to rotate according to the position of the target image block in the target image, and trigger the acquisition unit 601 to continue to acquire the target image taken by the target inspection robot through the camera until at least one valve hall device is identified.

[0150] It can be seen that by implementing the device described in the above embodiment, the position and posture of the gimbal are adjusted back and forth multiple times, so that the offset of the gimbal can be reduced to a smaller threshold, thereby realizing accurate calibration of the position and posture of the robot gimbal, and accurately restoring it to the preset position, which is beneficial to eliminating the adverse effects of the accumulated offset on the shooting target device; at the same time, it can also improve the fault tolerance of the gimbal calibration, making the gimbal calibration more flexible and reliable.

[0151] See also Figure 7 , Figure 7 Schematic diagram of a modular electronic device disclosed in an embodiment of the present application. Figure 7As shown, the electronic device may include:

[0152] A memory 701 storing executable program codes;

[0153] a processor 702 coupled to the memory 701;

[0154] The processor 702 calls the executable program code stored in the memory 701 to execute all or part of the steps in any one of the pan / tilt calibration methods for the valve hall equipment inspection robot described in the above embodiments.

[0155] The memory 701 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 701 includes a non-transitory computer-readable storage medium. The memory 701 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 701 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data created according to the use of the server, etc.

[0156] The processor 702 may include one or more processing cores. The processor 702 uses various interfaces and lines to connect various parts of the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 701, and calling data stored in the memory 701. Optionally, the processor 702 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 702 can integrate one or more combinations of a central processing unit (CPU) and a modem. Among them, the CPU mainly processes operating systems and applications, etc.; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 702, but may be implemented separately through a chip.

[0157] In addition, an embodiment of the present application further discloses a computer-readable storage medium, which stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute all or part of the steps in any of the pan-tilt calibration methods for a valve hall equipment inspection robot described in the above embodiments.

[0158] In addition, an embodiment of the present application further discloses a computer program product. When the computer program product is run on a computer, the computer can execute all or part of the steps in any of the pan-tilt calibration methods for the valve hall equipment inspection robot described in the above embodiments.

[0159] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0160] The above is a detailed introduction to the pan-tilt calibration method, device and storage medium of a valve hall equipment inspection robot disclosed in an embodiment of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A pan / tilt calibration method for a valve hall equipment inspection robot, characterized in that: include: When the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture, a target image captured by the camera of the target inspection robot is obtained; Identify the valve hall device in the target image to obtain the identified target device; determining image coordinates of the target device in the target image; Acquire preset coordinates corresponding to the target device from a database, and calculate a coordinate offset according to the preset coordinates and the image coordinates; Calculating a position adjustment amount and a posture adjustment amount for the gimbal according to the coordinate offset; Adjusting the preset position according to the position adjustment amount, and / or adjusting the preset posture according to the posture adjustment amount; Adjusting the preset position according to the position adjustment amount, or adjusting the preset posture according to the posture adjustment amount, includes: Determining whether the posture adjustment amount is greater than a preset adjustment threshold; When the amount is greater than the preset adjustment threshold, a first moving direction and a first moving distance corresponding to the position adjustment amount are obtained, and the gimbal is controlled to move the first moving distance toward the first moving direction, and the preset position is updated according to the position of the gimbal after the movement; When the amount is not greater than the preset adjustment threshold, obtaining a first rotation direction and a first rotation angle corresponding to the posture adjustment amount, and controlling the pan / tilt to rotate so that the camera rotates toward the first rotation direction by the first rotation angle, and updating the preset posture according to the posture of the pan / tilt after rotation; After calculating the position adjustment amount and attitude adjustment amount for the gimbal, the method further includes: obtaining the historical adjustment data of the gimbal from the database of the host computer, the historical adjustment data including the position adjustment amount record, attitude adjustment amount record and adjustment times of the gimbal; and correcting the position adjustment amount and the attitude adjustment amount according to the historical adjustment data.

2. The method according to claim 1, characterized in that: The preset position is adjusted according to the position adjustment amount, and the preset posture is adjusted according to the posture adjustment amount, including: Acquire a second moving direction and a second moving distance corresponding to the position adjustment amount, and control the pan / tilt head to move the second moving distance toward the second moving direction; Acquire a second rotation direction and a second rotation angle corresponding to the posture adjustment amount, and control the pan / tilt head to rotate so that the camera rotates toward the second rotation direction and the second rotation angle; The preset position is updated according to the position of the pan-tilt platform after movement, and the preset posture is updated according to the posture of the pan-tilt platform after rotation.

3. The method according to claim 1, characterized in that The adjusting the preset position according to the position adjustment amount, and / or adjusting the preset posture according to the posture adjustment amount, includes: When the coordinate offset is greater than a preset deviation threshold, the preset position is adjusted according to the position adjustment amount, and / or the preset posture is adjusted according to the posture adjustment amount, and the process of obtaining the target image captured by the target inspection robot through the camera is continued until the coordinate offset is no greater than the preset deviation threshold.

4. The method according to claim 3, characterized in that After adjusting the preset position according to the position adjustment amount, and / or adjusting the preset posture according to the posture adjustment amount, and continuing to obtain the target image captured by the target inspection robot through the camera until the coordinate offset is not greater than the preset deviation threshold, the method further includes: Acquire multiple groups of adjustment data calculated in the process of adjusting the preset position and the preset posture, each group of adjustment data includes the coordinate offset, the position adjustment and the posture adjustment calculated each time; When the number of the plurality of groups of adjustment data is greater than a preset group number threshold, mapping relationships between the coordinate offset and the position adjustment and the posture adjustment are fitted according to the plurality of groups of adjustment data; The mapping relationship is stored in a database.

5. The method according to claim 1, characterized in that The step of identifying the valve hall device in the target image to obtain the identified target device includes: According to the valve hall equipment characteristic data stored in the database, identifying one or more valve hall equipment in the target image that matches the valve hall equipment characteristic data; The distances between the one or more valve hall devices and the center of the target image in the target image are calculated, and a valve hall device that is closest to the center of the target image is used as the identified target device.

6. The method according to claim 5, characterized in that When it is identified that there is no valve hall equipment in the target image according to the valve hall equipment feature data stored in the database, the method further includes: Identify whether there are one or more suspected image blocks in the target image, wherein feature data of the suspected image blocks partially matches feature data of the valve hall device; Acquire one with the highest matching degree from the one or more suspected image blocks as the target image block; According to the position of the target image block in the target image, the pan / tilt is controlled to drive the camera to rotate, and the step of obtaining the target image captured by the target inspection robot through the camera is continued until at least one valve hall device is identified.

7. A pan / tilt calibration device for a valve hall equipment inspection robot, characterized in that: include: An acquisition unit, configured to acquire a target image captured by a camera of the target inspection robot when the pan-tilt platform connected to the target inspection robot reaches a preset position and the pan-tilt platform is in a preset posture; An identification unit, used for identifying the valve hall device in the target image to obtain the identified target device; a determining unit, configured to determine the image coordinates of the target device in the target image; A first calculation unit, configured to obtain preset coordinates corresponding to the target device from a database, and calculate a coordinate offset according to the preset coordinates and the image coordinates; A second calculation unit, used for calculating a position adjustment amount and a posture adjustment amount for the gimbal according to the coordinate offset; an adjusting unit, configured to adjust the preset position according to the position adjustment amount, and / or adjust the preset posture according to the posture adjustment amount; The adjustment unit includes a judgment subunit and a first adjustment subunit: The judging subunit is used to judge whether the posture adjustment amount is greater than a preset adjustment threshold; The first adjustment subunit is used to obtain a first moving direction and a first moving distance corresponding to the position adjustment amount when the judgment subunit judges that it is greater than the preset adjustment threshold, and control the pan / tilt to move the first moving distance toward the first moving direction, and update the preset position according to the position of the pan / tilt after the movement; when the judgment subunit judges that it is not greater than the preset adjustment threshold, obtain a first rotation direction and a first rotation angle corresponding to the posture adjustment amount, and control the pan / tilt to rotate so that the camera rotates the first rotation angle toward the first rotation direction, and update the preset posture according to the posture of the pan / tilt after the rotation; The apparatus also includes means for performing the following operations: After the second calculation unit calculates the position adjustment amount and the attitude adjustment amount for the gimbal, the historical adjustment data of the gimbal is obtained from the database of the host computer, and the historical adjustment data includes the position adjustment amount record, the attitude adjustment amount record and the number of adjustments of the gimbal; the position adjustment amount and the attitude adjustment amount are corrected according to the historical adjustment data.

8. An electronic device, characterized in that: include: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the method according to any one of claims 1 to 6.

Images