Unmanned aerial vehicle speed measurement method for high-voltage transmission line in strong electromagnetic field environment and related device

Abstract

The invention provides an unmanned aerial vehicle speed measurement method and a related device for a high-voltage transmission line in a strong electromagnetic field environment. The method comprises the following steps: extracting ORB feature points of each frame of image from continuous frames of images shot by an unmanned aerial vehicle by adopting an ORB algorithm; performing rough matching on the ORB feature points of any two adjacent frames of images based on a first matching condition to obtain a matching point pair relation set of the two adjacent frames of images; performing clustering processing on respective feature points of two adjacent frames of images in the matching point pair relation set to obtain respective cluster sets of the two adjacent frames of images; performing fine matching on the feature points between the two cluster sets based on a second matching condition to obtain two updated cluster sets; and based on the matching point pairs and the related clusters, calculating the moving speed of each cluster by using an optical flow method, and taking the moving speed after fusion processing as the speed of the unmanned aerial vehicle. Through the image processing technology, the system structure is simplified, the measurement precision is improved, and the cruising ability and the overall performance of the unmanned aerial vehicle are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle monitoring, and in particular relates to a method and a related device for measuring the speed of an unmanned aerial vehicle in a strong electromagnetic field environment of a high-voltage transmission line. Background Art

[0002] In the southern hilly areas, overhead power transmission lines are usually laid in mountains and forests and inaccessible places. The positioning signals in these places are weaker than those in urban areas. At the same time, there is a strong electromagnetic field environment near the high-voltage overhead power transmission lines. The strong electromagnetic field has a strong interference on the communication control signal between the remote control and the drone. The drone's own sensors are affected when it approaches the overhead power transmission lines, thereby reducing accuracy or failure. When the drone is equipped with an X-ray machine to shoot near the tension section of the transmission line, it is often necessary for the drone to stay for a period of time and try to stay as close to the tension clamp as stably as possible. In this process, it is necessary to ensure that various sensors, such as magnetometers and gyroscopes, are less affected by the electromagnetic environment to ensure the stability and safety of the drone and the X-ray machine. There are many methods to solve these problems, such as the method of using a compensator, by measuring the magnetic strength of the interference magnetic field around the drone, and then offsetting the influence of electromagnetic interference by generating a magnetic field, or installing an electromagnetic interference protection device for the drone outside the drone body to isolate electromagnetic interference.

[0003] Existing technical methods, such as the electromagnetic field compensation method, require the additional installation of electromagnetic measurement and generation equipment. The electromagnetic generation equipment itself will generate certain current harmonics, which will interfere with the operation of the sensor. In addition, when it is necessary to accurately measure the phase of the electromagnetic field, the instrument accuracy may not meet the requirements. The method of installing an electromagnetic interference protection device requires additional load for drones that already have a X-ray machine, which will significantly reduce the drone's flight time. Summary of the invention

[0004] In view of this, the present invention provides a method and related devices for measuring the speed of a UAV in a strong electromagnetic field environment of a high-voltage transmission line, so as to solve the above-mentioned problems existing in the existing speed measurement methods in a strong electromagnetic environment.

[0005] In order to achieve the above object, the technical solution provided by the present invention is as follows:

[0006] In a first aspect, the present invention provides a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line, comprising the following steps:

[0007] The ORB algorithm is used to extract the ORB feature points of each frame from the continuous frame images taken by the drone;

[0008] Roughly matching the ORB feature points of any two adjacent frames of images based on a first matching condition to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each ORB feature point of the previous frame of image and the matching ORB feature point in the next frame of image satisfy a preset first position proximity relationship;

[0009] Performing clustering processing on the feature points of two adjacent frames of images in the matching point pair relationship set to obtain the cluster sets of the two adjacent frames of images;

[0010] Based on the second matching condition, the feature points between the two cluster sets are precisely matched to obtain two updated cluster sets, wherein the second matching condition is that in two related clusters between the two cluster sets, all feature points have their own matching point pairs and the two related clusters satisfy a preset second position proximity relationship;

[0011] Based on the matching point pairs and related clusters, the optical flow method is used to calculate the moving speed of each cluster, the moving speeds of all clusters are fused, and the fused moving speed is used as the speed of the drone.

[0012] Furthermore, based on the first matching condition, the ORB feature points of any two adjacent frames of images are roughly matched to obtain a set of matching point pair relationships of the two adjacent frames of images, including:

[0013] Each ORB feature point in the previous frame image is used as a template point, and the K nearest neighbor algorithm is used to match K ORB feature points as matching points for each template point in the adjacent next frame image. A matching point pair relationship set is formed based on the template points and the matching points. K is a preset value, and the first position proximity relationship is the position proximity relationship satisfied by the matching point pairs matched by the K nearest neighbor algorithm.

[0014] Furthermore, after obtaining the set of matching point pair relationships between two adjacent frames of images, the method further includes:

[0015] Calculate the Hamming distance between each matching point and the template point;

[0016] All matching point pairs whose ratio of the Hamming distance of the nearest neighbor matching point to the Hamming distance of the second nearest neighbor matching point is greater than the set threshold are eliminated to obtain a new matching point pair relationship set.

[0017] Furthermore, based on the second matching condition, the feature points between the two cluster sets are precisely matched to obtain two updated cluster sets, including:

[0018] Calculate the pixel distance between each two cluster centroids between two cluster sets;

[0019] For each set of matching point pairs in the matching point pair relationship set, if any point in the set is not in the cluster set and / or the pixel distance between the two related clusters corresponding to the matching point pair does not meet the set threshold, the set of matching point pairs is deleted and the corresponding two related clusters are updated;

[0020] Based on the finally updated related clusters, two updated cluster sets are obtained, and the second location proximity relationship is the location proximity relationship satisfied between the finally updated related clusters.

[0021] Furthermore, based on the matching point pairs and related clusters, the moving speed of each cluster is calculated using the optical flow method, including:

[0022] For any group of related clusters, establish the optical flow equations of all matching point pairs in the related clusters, and the number of matching point pairs is at least 2;

[0023] Solve the optical flow equation and convert the obtained speed into the movement speed of the current cluster.

[0024] Furthermore, the moving speeds of all clusters are fused, including:

[0025] The moving speeds of all clusters are treated as outliers;

[0026] The remaining moving speeds after outliers are processed are averaged.

[0027] Furthermore, after the fused moving speed is used as the speed of the drone, the following is also included:

[0028] The fused moving speed is converted into the speed in the world coordinate system to facilitate the servo control of the UAV motor speed.

[0029] In a second aspect, the present invention provides a drone speed measurement device in a strong electromagnetic field environment of a high-voltage transmission line, comprising:

[0030] A feature point detection module is used to extract the ORB feature points of each frame image from the continuous frame images taken by the drone using the ORB algorithm;

[0031] A coarse matching module is used to perform coarse matching on the ORB feature points of any two adjacent frames of images based on a first matching condition to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each ORB feature point of the previous frame of image satisfies a preset first position proximity relationship with the matching ORB feature point in the next frame of image;

[0032] A precise matching module is used to perform clustering processing on the feature points of each of the two adjacent frames of images in the matching point pair relationship set to obtain the cluster sets of each of the two adjacent frames of images; and to perform precise matching on the feature points between the two cluster sets based on a second matching condition to obtain two updated cluster sets, wherein the second matching condition is that in two related clusters between the two cluster sets, all feature points have their own matching point pairs and the two related clusters satisfy a preset second position proximity relationship;

[0033] The speed calculation module is used to calculate the moving speed of each cluster based on the matching point pairs and related clusters using the optical flow method, fuse the moving speeds of all clusters, and use the fused moving speed as the speed of the drone.

[0034] In a third aspect, the present invention further provides a computer device, the device comprising a processor and a memory:

[0035] The memory is used to store the computer program and send the instructions of the computer program to the processor;

[0036] The processor executes a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to the instructions of the computer program as described in the first aspect.

[0037] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to the first aspect is implemented.

[0038] In summary, the present invention provides a method and related devices for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line, including extracting ORB feature points of each frame of image from continuous frame images taken by a drone using an ORB algorithm; roughly matching the ORB feature points of any two adjacent frames of image based on a first matching condition to obtain a matching point pair relationship set of the two adjacent frames of image, the first matching condition being that each ORB feature point of the previous frame of image and the matching ORB feature point in the next frame of image satisfy a preset first position proximity relationship; clustering the feature points of each of the two adjacent frames of image in the matching point pair relationship set to obtain cluster sets of each of the two adjacent frames of image; precisely matching the feature points between the two cluster sets based on a second matching condition to obtain two updated cluster sets, the second matching condition being that in two related clusters between the two cluster sets, all feature points have their own matching point pairs and the two related clusters satisfy a preset second position proximity relationship; based on the matching point pairs and the related clusters, calculating the moving speed of each cluster using an optical flow method, fusing the moving speeds of all clusters, and using the fused moving speed as the speed of the drone. The present invention solves the problem of UAV speed measurement in the strong electromagnetic field environment of high-voltage transmission lines in the prior art through image processing technology, which not only simplifies the system structure and improves the measurement accuracy, but also significantly improves the endurance and overall performance of the UAV. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A flowchart of a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line provided by an embodiment of the present invention;

[0041] Figure 2 A block diagram of a device for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line provided by an embodiment of the present invention;

[0042] Figure 3 A block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 This embodiment provides a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line, comprising the following steps:

[0045] S11: The ORB algorithm is used to extract the ORB feature points of each frame image from the continuous frame images taken by the drone.

[0046] It should be noted that the ORB algorithm (Oriented FAST and Rotated BRIEF) is an efficient feature point detection and description method. It combines the speed advantage of the FAST corner detector and the simplicity of the BRIEF feature descriptor, and provides rotation invariance by adding directional information.

[0047] The specific feature extraction steps are as follows:

[0048] (1) Feature point detection: Use the FAST corner detector to preliminarily find feature points in the image.

[0049] (2) Calculate the direction of the feature point: For each key point, calculate the centroid of the pixels around it, and then calculate the vector from the feature point to the centroid. The angle of this vector is the direction of the feature point. Specifically, if the point in the image is , whose intensity centroid is C, then the angle It can be calculated by the following formula:

[0050]

[0051] in, are the pixels around the feature point, C is the centroid of these pixels, and x and y are the coordinates of the pixels.

[0052] (3) When generating the BRIEF descriptor, ORB takes into account the direction of the feature point. Specifically, it rotates the compared pixel pairs according to the direction of the feature point, and then compares their grayscale values. The BRIEF descriptor generated in this way is an ORB feature point with rotation invariance.

[0053] S12: Based on the first matching condition, the ORB feature points of any two adjacent frames of images are roughly matched to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each ORB feature point of the previous frame of image and the matching ORB feature point in the next frame of image satisfy a preset first position proximity relationship.

[0054] It should be noted that the ORB feature points between two adjacent frames are preliminarily matched through the first matching condition. The first matching condition here is based on the position proximity relationship, that is, each ORB feature point in the previous frame tries to find a corresponding point with a similar position in the next frame. The data set formed after the rough matching process and containing the correspondence relationship between the ORB feature points of the previous and next frames is the matching point pair relationship set.

[0055] S13 performs clustering processing on the feature points of two adjacent frames of images in the matching point pair relationship set to obtain cluster sets of two adjacent frames of images.

[0056] It should be noted that clustering is an unsupervised learning method that aims to divide data into different groups (clusters) so that members in the same group have high similarity to each other, while members in different groups have low similarity. In this step, for the previously obtained set of matching point pair relationships, ORB feature points that are related to each other in the same frame are grouped or clustered.

[0057] S14: Based on the second matching condition, the feature points between the two cluster sets are precisely matched to obtain two updated cluster sets. The second matching condition is that in the two related clusters between the two cluster sets, all feature points have their own matching point pairs and the two related clusters satisfy a preset second position proximity relationship.

[0058] It should be noted that this step is a process to further improve the matching accuracy. The second matching condition requires that not only individual feature points can be matched in pairs, but also the corresponding clusters in different frames to which these points belong must satisfy a certain position proximity relationship. This helps to filter out false matches and improve overall accuracy.

[0059] S15: Based on the matching point pairs and related clusters, the moving speed of each cluster is calculated using the optical flow method, the moving speeds of all clusters are fused, and the fused moving speed is used as the speed of the drone.

[0060] It should be noted that the optical flow method is a technique used to estimate the motion of objects in video sequences. The direction and speed of an object's movement are inferred by analyzing the changes in pixels over time. In this step, the optical flow method is used to estimate the actual moving speed of each cluster (representing one or more feature points) based on the known matching point pairs and their corresponding cluster information. Finally, the speed information of all clusters is fused in some way to obtain the overall motion speed estimate of the drone in the entire scene.

[0061] According to the actual situation of the drone working on site, the inventor of this application has the following conclusions about the relationship between the continuous frames taken by the drone's lower camera: (1) The relative movement between the drone and the ground is the movement of the drone relative to the overall stationary ground, which meets the "spatial consistency" assumption of the Lucas-Kanade optical flow method; (2) The overhead power transmission lines at the work site are often erected on mountains, wastelands, and water bodies. The overall color features of these regional scenes in the image are not rich enough. The method based on color features cannot capture the local feature information in the image. The color feature descriptor is unstable or the matching effect is not good, so it is necessary to introduce ORB features to enhance the expression of local spatial texture information in the image. Based on the above conclusions, this embodiment proposes a strategy based on ORB features and multi-level feature screening to improve the matching accuracy of matching points, and successfully screens out valuable matching points through the multi-level screening strategy. Finally, based on the screened matching point set, the calculation formula of the Lucas-Kanade optical flow method is used to calculate the common moving speed of these clusters, and the actual moving speed of the drone is inferred. The method for measuring the speed of a UAV in a strong electromagnetic field environment of a high-voltage transmission line provided in this embodiment can effectively eliminate mismatching and redundant matching information, retain high-quality matching point pairs, and better solve the problem of loss of UAV speed data when electromagnetic field interference occurs.

[0062] In one embodiment, based on the first matching condition, ORB feature points of any two adjacent frames of images are roughly matched to obtain a set of matching point pair relationships of the two adjacent frames of images, including:

[0063] S21: Taking each ORB feature point in the previous frame image as a template point, using the K nearest neighbor algorithm to match K ORB feature points as matching points for each template point in the adjacent next frame image, forming a matching point pair relationship set based on the template points and the matching points, K is a preset value, and the first position proximity relationship is the position proximity relationship satisfied by the matching point pairs matched by the K nearest neighbor algorithm.

[0064] It should be noted that, taking the ORB feature points obtained in the above steps as an example, when using the above method to analyze the ground image taken by the drone's bottom camera ( , for any number of adjacent frames , After feature extraction, the image The set of all feature points is: :{ , ,…},picture The set of all feature points is: { , , …}. For the collection , , use the KNN algorithm (taking K=2 as an example) to quickly perform rough matching and establish a one-to-two matching sequence, that is, for Any template element in The two matching elements of the nearest neighbor and the second nearest neighbor can be found to form a matching point pair relationship set .

[0065] In the above matching process, the first position proximity relationship refers to the degree of proximity in spatial position between the matching point pairs determined based on the KNN algorithm. That is, only when the distance between an ORB feature point in the i-th frame and two feature points in the i+1-th frame is small enough to meet the preset threshold condition, they are considered to constitute a valid matching pair.

[0066] In a further embodiment, after obtaining the set of matching point pair relationships between two adjacent frames of images, the method further includes:

[0067] S31: Calculate the Hamming distance between each matching point and the template point.

[0068] S32: Eliminate all matching point pairs whose ratios of the Hamming distance of the nearest neighbor matching point to the Hamming distance of the second nearest neighbor matching point are greater than a set threshold, and obtain a new matching point pair relationship set.

[0069] It should be noted that after obtaining the matching point pair relationship set, the matching point pair set is calculated. The Hamming distance between the template point and the matching point. Since the closer the distance, the higher the reliability of the match, when all points are close, the reliability of the match is not high, and when the nearest neighbor matching point is close and other points are relatively far away, it can be considered that the reliability of the match is greatly increased, that is, based on the distance characteristics between the correct match and the wrong match, a more reliable correct match relationship can be screened out. Based on this, the threshold can be set. , roughly eliminate all the nearest neighbor points whose ratio of the Hamming distance to the second nearest neighbor point is greater than the threshold The matching point pairs are obtained to obtain a new matching point pair relationship set .

[0070] In one embodiment, the feature points between the two cluster sets are precisely matched based on the second matching condition to obtain two updated cluster sets, including:

[0071] S41: Calculate the pixel distance between every two cluster centroids between two cluster sets.

[0072] S42: For each group of matching point pairs in the matching point pair relationship set, if any point in the group is not in the cluster set and / or the pixel distance between the two related clusters corresponding to the matching point pair does not meet the set threshold, the group of matching point pairs is deleted and the corresponding two related clusters are updated.

[0073] S43: obtaining two updated cluster sets based on the finally updated related clusters, wherein the second location proximity relationship is a location proximity relationship satisfied between the finally updated related clusters.

[0074] It should be noted that the current matching point pair relationship set There are still some wrong matching relationships that need to be eliminated. Since the images involved in this patent are taken by the drone's bottom camera, the template image taken during the registration and matching images There is no huge change in the three-dimensional space, such as flipping and large-scale rotation, etc. The two almost only have a translation relationship on the two-dimensional plane, so the captured image has excellent smoothness, that is, when the feature point moves, its neighboring pixels will move together in the same direction. According to the above situation analysis and the corresponding relationship judgment when the image feature points are aligned: in the matching point alignment relationship, there are usually more matching points near the successfully matched points, while the matching failed points lack such a corresponding relationship. In this regard, this embodiment proposes a feature matching method based on matching point clustering, which does not need to set the image to grid for algorithm acceleration like the method based on motion statistics, and then judges the number of matching points in the grid as the basis for supporting the success or failure of the match.

[0075] This method firstly sets the matching point pair relationship set after the initial screening. The template points and matching points in the clustering operation are clustered separately. Here, the Mean-Shift clustering method can be selected. The kernel function radius of the clustering method should not be set too small to avoid erroneously excluding the correct matching points. Two cluster sets are obtained: Cluster set and cluster sets , for any cluster or , which contains the set or For several aggregation points in , For all clusters in the cluster, calculate the pixel distance between their centroids one by one ). The next step is to further eliminate the coarse screening matching points based on the above analysis and calculation. The elimination is mainly based on two principles: (1) ), remove Not Available or Not Available The matching point in the cluster will be deleted as long as one of the conditions is met; (2) ,like and At the same time , On a cluster of , but the centroid pixel distance between the two clusters ) is greater than the threshold ( The camera frame rate , the maximum moving speed of the drone And the conversion coefficient between the pixel coordinate system and the actual coordinate system After the above fine screening, the matching relationship of the points that meet the requirements is obtained. With the point matching relationship, according to the experience of image registration, it can be known that: compared with the unrelated or partially related cluster areas, the mutually related cluster areas often have the most point matching relationships. Therefore, S groups of one-to-one corresponding clusters are screened out to form a set of one-to-one corresponding relationships of related clusters, that is, | )}.

[0076] In one embodiment, based on the matching point pairs and the related clusters, the moving speed of each cluster is calculated using the optical flow method, including:

[0077] S51: for any group of related clusters, establish optical flow equations for all matching point pairs in the related clusters, and the number of matching point pairs is at least not less than 2;

[0078] S52: Solve the optical flow equation, and convert the obtained speed into the moving speed of the current cluster.

[0079] It should be noted that for the matching points and the interrelated clusters after the fine screening operation, the moving speed of each cluster can be calculated in turn. According to the derivation of the optical flow method, the light intensity of any pixel on the cluster through Move after time After the distance, arrive , since the light intensity of the same pixel in the previous and next frames of the image remains unchanged, we have ,Right now ( , , Respectively represent the light intensity of the pixel in the image in the X, Y direction and time The partial derivative on , is the speed of the pixel point moving in the X and Y axis directions), a constraint equation with two unknowns, the equation has no definite solution, so the assumption is introduced: when the feature point moves, its neighboring pixels will move in the same direction together, that is, all points in the same cluster have the same , Substituting all the points on the cluster, we get the system of equations:

[0080]

[0081] Combining the above equations, we get The calculation formula is:

[0082] =

[0083] The calculation method for other clusters is the same as above, and the speed of each cluster in the X and Y directions can be obtained , ( = ), merge the speeds of each cluster in the X and Y directions to get the true speed , which is a velocity vector with magnitude and direction, there are S of them, recorded as the set }.

[0084] In one embodiment, the moving speeds of all clusters are fused, including:

[0085] S61: Process the moving speeds of all clusters as outliers;

[0086] S62: averaging the remaining moving speeds after the outliers are processed.

[0087] It should be noted that the RANSAC method can be used to calculate the speed of the set V. and direction Perform outlier processing (if any of the magnitude or direction is abnormal, the velocity vector is excluded). , in size and direction Take the average value of is the measured speed of the UAV in the pixel coordinate system.

[0088] In one embodiment, if the measured speed is to be used to implement servo control of the motor speed of the drone, the speed measured by the patented method must be converted into a speed in the world coordinate system.

[0089] Based on the same inventive concept, the embodiment of the present application also provides a device for measuring the speed of a drone under a strong electromagnetic field environment of a high-voltage transmission line, which is used to implement the method for measuring the speed of a drone under a strong electromagnetic field environment of a high-voltage transmission line. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiment of the device for measuring the speed of a drone under a strong electromagnetic field environment of a high-voltage transmission line provided below can be referred to the limitations of the method for measuring the speed of a drone under a strong electromagnetic field environment of a high-voltage transmission line above, and will not be repeated here.

[0090] See also Figure 2 This embodiment provides a UAV speed measurement device in a high-voltage transmission line strong electromagnetic field environment, including:

[0091] A feature point detection module is used to extract the ORB feature points of each frame image from the continuous frame images taken by the drone using the ORB algorithm;

[0092] A coarse matching module is used to perform coarse matching on the ORB feature points of any two adjacent frames of images based on a first matching condition to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each ORB feature point of the previous frame of image satisfies a preset first position proximity relationship with the matching ORB feature point in the next frame of image;

[0093] A precise matching module is used to perform clustering processing on the feature points of each of the two adjacent frames of images in the matching point pair relationship set to obtain the cluster sets of each of the two adjacent frames of images; and to perform precise matching on the feature points between the two cluster sets based on a second matching condition to obtain two updated cluster sets, wherein the second matching condition is that in two related clusters between the two cluster sets, all feature points have their own matching point pairs and the two related clusters satisfy a preset second position proximity relationship;

[0094] The speed calculation module is used to calculate the moving speed of each cluster based on the matching point pairs and related clusters using the optical flow method, fuse the moving speeds of all clusters, and use the fused moving speed as the speed of the drone.

[0095] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0096] Reference Figure 3 An embodiment of the present invention further provides a computer device, including: a memory and a processor and a computer program stored in the memory. When the computer program is executed on the processor, the method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line as described in any one of the above methods is implemented.

[0097] The computer device may be a desktop computer, a notebook, a PDA, a cloud server or other computing device. The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Figure 3 It is only an example of a computer device and does not constitute a limitation of the computer device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0098] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0099] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device. Further, the memory may include both an internal storage unit and an external storage device of the computer device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been output or is to be output.

[0100] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line as described in any one of the above methods is implemented.

[0101] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, USB flash drive, mobile hard disk, disk or optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0102] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0103] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0104] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0105] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line, characterized in that: The steps include: The ORB algorithm is used to extract the ORB feature points of each frame from the continuous frame images taken by the drone; The ORB feature points of any two adjacent frames of images are roughly matched based on a first matching condition to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each of the ORB feature points of the previous frame of image and the matched ORB feature points in the next frame of image satisfy a preset first position proximity relationship; Performing clustering processing on the feature points of two adjacent frames of images in the matching point pair relationship set to obtain cluster sets of two adjacent frames of images; Based on a second matching condition, the feature points between the two cluster sets are precisely matched to obtain the two updated cluster sets, wherein the second matching condition is that in two related clusters between the two cluster sets, all feature points have respective matching point pairs and the two related clusters satisfy a preset second position proximity relationship; Based on the matching point pairs and the related clusters, the moving speed of each cluster is calculated using the optical flow method, the moving speeds of all clusters are fused, and the fused moving speed is used as the speed of the drone.

2. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 1 is characterized in that: Based on the first matching condition, the ORB feature points of any two adjacent frames of images are roughly matched to obtain a set of matching point pair relationships of the two adjacent frames of images, including: Each of the ORB feature points in the previous frame image is used as a template point, and a K nearest neighbor algorithm is used to match K ORB feature points as matching points for each of the template points in an adjacent subsequent frame image. A matching point pair relationship set is formed based on the template points and the matching points, K is a preset value, and the first position proximity relationship is a position proximity relationship satisfied by the matching point pairs matched by the K nearest neighbor algorithm.

3. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 2 is characterized in that: After obtaining the set of matching point pairs between two adjacent frames of images, the following steps are also included: Calculating the Hamming distance between each matching point and the template point; All matching point pairs whose ratio of the Hamming distance of the nearest neighbor matching point to the Hamming distance of the second nearest neighbor matching point is greater than a set threshold are eliminated to obtain a new set of matching point pair relationships.

4. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 1 is characterized in that: Based on the second matching condition, the feature points between the two cluster sets are precisely matched to obtain the two updated cluster sets, including: Calculating the pixel distance between every two cluster centroids between two cluster sets; For each group of matching point pairs in the matching point pair relationship set, if any point in the group is not in the cluster set and / or the pixel distance between two related clusters corresponding to the matching point pair does not meet a set threshold, the matching point pair in the group is deleted, and the corresponding two related clusters are updated; The two updated cluster sets are obtained based on the finally updated related clusters, and the second location proximity relationship is a location proximity relationship satisfied between the finally updated related clusters.

5. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 1 is characterized in that: Based on the matching point pairs and the related clusters, the moving speed of each cluster is calculated using an optical flow method, including: For any group of the related clusters, establish the optical flow equations of all the matching point pairs in the related clusters, and the number of the matching point pairs is at least not less than 2; The optical flow equation is solved, and the obtained speed is converted into the moving speed of the current cluster.

6. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 1 is characterized in that: The moving speeds of all clusters are fused, including: Perform outlier processing on the moving speeds of all clusters; The moving speeds remaining after outlier processing are averaged.

7. The method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to claim 1 is characterized in that: After the fused moving speed is used as the speed of the drone, the method further includes: The fused moving speed is converted into a speed in the world coordinate system to facilitate servo control of the motor speed of the drone.

8. A drone speed measurement device in a high-voltage transmission line strong electromagnetic field environment, characterized in that: include: A feature point detection module is used to extract the ORB feature points of each frame image from the continuous frame images taken by the drone using the ORB algorithm; A coarse matching module, configured to perform coarse matching on the ORB feature points of any two adjacent frames of images based on a first matching condition to obtain a set of matching point pair relationships between the two adjacent frames of images, wherein the first matching condition is that each of the ORB feature points of the previous frame of image and the matching ORB feature points in the next frame of image satisfy a preset first position proximity relationship; A precise matching module, used for clustering the feature points of two adjacent frames of images in the matching point pair relationship set to obtain cluster sets of two adjacent frames of images; Furthermore, based on a second matching condition, the feature points between the two cluster sets are precisely matched to obtain the two updated cluster sets, wherein the second matching condition is that in two related clusters between the two cluster sets, all feature points have respective matching point pairs and the two related clusters satisfy a preset second position proximity relationship; The speed calculation module is used to calculate the moving speed of each cluster based on the matching point pairs and the related clusters using the optical flow method, fuse the moving speeds of all clusters, and use the fused moving speeds as the speed of the drone.

9. A computer device, characterized in that: The device comprises a processor and a memory: The memory is used to store a computer program and send instructions of the computer program to the processor; The processor executes a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line according to the instructions of the computer program as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, a method for measuring the speed of a drone in a strong electromagnetic field environment of a high-voltage transmission line is implemented as described in any one of claims 1-7.

Images