A system, device and method for correcting bright and dark lines of LED display screen by unmanned aerial vehicle

Abstract

The present invention relates to a system, device and method for unmanned aerial vehicle (UAV) to correct the light and dark lines of LED display screens. The method comprises selecting a hovering position; determining the longitudinal relative position of the UAV and the ground; planning the flight route of the UAV; determining the starting position of the UAV flight route through an OCR recognition visual perception unit; sequentially identifying the digital label of each module of the LED display screen; sequentially photographing the modules one by one to form a photographing path, and then transmitting the high-definition images of the modules obtained by sequential photographing to a display screen control subsystem in real time by the UAV; outputting the corresponding light and dark line correction coefficient for the high-definition image corresponding to each module, and then directly feeding back the light and dark line correction coefficient to the display screen light and dark line correction unit to realize the light and dark line correction of each module. The method has the beneficial effect of being able to facilitate the massive calculation of the correction coefficient in the later stage and improving the efficiency and accuracy of LED display screen proofreading.

Description

Technical Field

[0001] The present invention relates to the field of LED display screen correction, and in particular to a system, device and method for unmanned aerial vehicle correction of bright and dark lines of an LED display screen. Background Art

[0002] With the popularization of LED display screens in various fields, from indoor all-in-one conference machines to outdoor exterior walls of high-rise buildings, as well as applications in airports, subways and other transportation hubs, LED display screens have been fully integrated into all aspects of people's lives and can bring people a wonderful visual experience.

[0003] The whole LED display is made up of small modules. Due to the structural differences of each module and the installation skills of the master, there will be bright and dark lines on the LED display. The reason for the bright lines is that the splicing between the modules is too tight; the reason for the dark lines is that the gap between the modules is too large. The generation of bright and dark lines will have a great impact on the overall display effect of the LED display, that is, the overall picture will be partially separated, which will affect the beauty of the complete picture.

[0004] The usual solution to the problem of bright and dark line correction is to debug the bright or dark lines of each cabinet in a "dark room" based on the visual experience of technicians, and then splice the debugged cabinets into a complete display screen, and debug the bright and dark lines between cabinets. This traditional debugging method is time-consuming and laborious. A more intelligent and applicable debugging method is needed, which can effectively eliminate the problem of bright and dark lines by directly debugging a complete display screen.

[0005] In view of the problems and limitations of traditional light and dark line debugging, this patent is based on OCR recognition algorithm and transplanted into the visual perception system of drones. It can plan the flight route according to our own needs to take pictures of each module, directly process the photos to obtain the light and dark line correction coefficients corresponding to each module, and then feed back the correction coefficients to each corresponding module, so as to achieve one-time light and dark line correction of the whole screen. The existing traditional debugging technology, from light and dark line debugging of each box to light and dark line debugging of the entire display screen, is time-consuming and laborious, and it completely relies on the visual experience of technicians to debug, and the accuracy and stability of light and dark line debugging are lacking. Summary of the invention

[0006] Based on this, it is necessary to provide a drone correction system for LED display screens to address the problem of an existing drone correction system for LED display screens.

[0007] An embodiment of the present application provides a drone system for correcting bright and dark lines of an LED display screen, the system comprising a drone control subsystem and a display screen control subsystem;

[0008] The drone control subsystem includes: an OCR recognition visual perception unit, a route planning module, and a module image acquisition and transmission unit; the drone control subsystem is arranged in a drone controller;

[0009] The display screen control subsystem comprises: a display screen bright and dark line correction unit; the display screen control subsystem is arranged in a display screen controller;

[0010] The OCR recognition visual perception unit includes a forward vision detection module, a forward vision OCR recognition module, a downward vision and ultrasonic sensor module, and a distance measurement perception module;

[0011] The forward vision detection module is used to detect the LED display screen directly in front of the drone and can select a horizontal fixed-point hovering position;

[0012] The lower visual and ultrasonic sensor modules are used to determine the longitudinal relative position of the drone with respect to the ground;

[0013] The forward vision OCR recognition module is used to determine the starting point of the UAV flight route through the OCR recognition visual perception unit, and sequentially identify the label in the center of each module of the LED display screen;

[0014] The route planning module is used to plan the flight route of the drone, using the starting point of the drone flight route as the origin of the drone flight, and setting a fixed horizontal moving distance and a vertical moving distance according to different models of LED display screens;

[0015] The module image acquisition and transmission unit is used to take photos of the modules one by one according to the route planning and the photo-taking sequence, and the module labels identified in sequence by the front vision OCR recognition module to form a photo-taking path, and then transmit the module high-definition images obtained by the sequential photos to the display screen control subsystem in real time by the drone;

[0016] The module image acquisition and transmission unit includes an acquisition module; the acquisition module is used to acquire the specified module image, and the acquisition field of view of the camera on the drone will include the module image adjacent to the four sides of the specified module, combined with the specified route planning and OCR recognition technology, to complete the flight mission of acquiring images;

[0017] The distance sensing module is used for the UAV to obtain the minimum safe sensing distance between the UAV and the LED display screen;

[0018] The display screen bright and dark line correction unit is used to output corresponding bright and dark line correction coefficients for the high-definition image corresponding to each module, and then directly feed back the bright and dark line correction coefficients to the display screen bright and dark line correction unit to achieve bright and dark line correction of each module.

[0019] An embodiment of the present invention also provides a method for calibrating bright and dark lines of an LED display screen by a drone, the method comprising:

[0020] Detect the LED display screen directly in front of the drone's vision and select a horizontal fixed-point hovering position;

[0021] Determine the longitudinal relative position of the drone to the ground;

[0022] Plan the flight route of the drone, use the starting point of the drone flight route as the origin of the drone flight, and set a fixed horizontal moving distance and vertical moving distance according to different models of LED display screens;

[0023] The starting point of the UAV flight route is determined by the OCR recognition visual perception unit; the label at the center of each module of the LED display screen is identified in turn; and according to the route planning and the order of taking photos, the module labels are sequentially identified by the front visual OCR recognition module to take photos of the modules one by one to form a photo path, and then the high-definition images of the modules obtained by the sequential photos are transmitted by the UAV to the display control subsystem in real time;

[0024] For each module's corresponding high-definition image, the corresponding light and dark line correction coefficient is output, and then the light and dark line correction coefficient is directly fed back to the display screen's light and dark line correction unit to achieve light and dark line correction for each module.

[0025] In some embodiments, the step of determining the starting point of the drone flight route by using an OCR recognition visual perception unit; and sequentially identifying the label at the center of each module of the LED display screen includes the following steps:

[0026] Obtain a labeled image dataset of LED display modules, select the LED display to test mode, and take photos of each module using a mobile phone, camera, or other shooting equipment. The requirements for taking photos require obtaining label information of the four sides of a single module and the center of the module.

[0027] By switching the interval between light points and the brightness value of light points, a characteristic and diverse image data set is obtained; for example, the characteristic and diverse image data set includes a total of 10,000 images, and the labels corresponding to each module have 10 different styles;

[0028] Obtain an OCR recognition model and select an OCR text recognition model combination; for example, select a CTPN+CRNN model combination with good OCR text recognition performance;

[0029] Retrain to obtain the corresponding model;

[0030] Transplant the OCR recognition model to the drone's OCR recognition visual perception unit.

[0031] In some embodiments, the step of planning the flight route of the drone includes:

[0032] Determine the coordinates of the return location and connect the drone handle via the mobile phone;

[0033] Get the starting coordinates of the flight route, and select the normal flight mode on the handle to make the drone quickly approach the LED display;

[0034] Carry out route planning and construct a route model for the flight route.

[0035] In some embodiments, the corresponding bright and dark line correction coefficient is output for the high-definition image corresponding to each module, and then the bright and dark line correction coefficient is directly fed back to the display screen bright and dark line correction unit, and the bright and dark line correction of each module is realized, including the steps of:

[0036] Calculate the correction coefficient of bright and dark lines of each module;

[0037] Compensation feedback correction coefficient; specifically, according to the order of module labels, when taking a picture of the current module, the compensation coefficient of the previous module is fed back to the LED display screen, and the cycle is repeated until the compensation of the correction coefficient of the last module is completed.

[0038] In some embodiments, the method further comprises:

[0039] Obtain the minimum safe perception distance between the drone and the LED display in real time;

[0040] If the safety sensing distance is less than the set value, an alarm is issued, and the drone is automatically made to fly parallel to the LED display screen at the minimum set value.

[0041] In some embodiments, the step of detecting the LED display screen directly in front of the drone vision further includes:

[0042] Set the joint seam between modules as the center line, and select two rows of lamp beads on both sides of the center line respectively;

[0043] Set any cross intersection between four adjacent modules as a non-lit light point.

[0044] In some embodiments, the module tags identified in sequence by the front vision OCR recognition module are photographed one by one in sequence to form a photographing path, including:

[0045] Take the leftmost module of the highest row of modules of the LED display screen as the flight origin of the drone, fly rightward along the highest row of modules to the rightmost module, then move down one row, fly leftward from the rightmost module to the leftmost module, then move down one row, and repeat the above steps.

[0046] In some embodiments, the step of obtaining a labeled image data set of an LED display screen module, selecting the LED display screen in a test mode, and photographing each module by a mobile phone, a camera, or other photographing equipment, wherein the photographing requirement requires obtaining label information of four sides of a single module and the center of the module, further comprises the steps of:

[0047] When collecting images of a specified module, the camera's field of view on the drone will include module images adjacent to the four sides of the specified module, and combined with the specified route planning and OCR recognition technology, the flight mission of collecting images is completed.

[0048] In some embodiments, the correction coefficient of the bright and dark lines of each module is calculated only once for the common edge formed by each module of the LED display screen connecting with other modules.

[0049] An embodiment of the present invention application further provides a device for correcting bright and dark lines of an LED display screen using a drone, including: a method for correcting bright and dark lines of an LED display screen using a drone as described in any embodiment of the present invention application.

[0050] An embodiment of the present application further provides a storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, implements a method for correcting bright and dark lines of an LED display screen by a drone according to any embodiment of the present application.

[0051] In a system for correcting the light and dark lines of an LED display screen provided by an unmanned aerial vehicle in an embodiment of the present application, when the LED display screen is in a test pattern mode, each module will have a corresponding digital number. Based on OCR recognition and the visual detection and positioning system of the unmanned aerial vehicle itself, the starting point of the unmanned aerial vehicle taking pictures of the LED display screen according to the route planning can be accurately determined, and then the fixed lateral movement distance and longitudinal movement distance are set according to different models of products to complete the photo taking of the entire display screen at one time. Through the transplantation of the OCR recognition algorithm, intelligent recognition can be achieved, the precise hovering position can be determined, and photos can be taken according to the specified route planning path to obtain photos corresponding to each module. The present invention application is a method for correcting the light and dark lines of an LED display screen by an unmanned aerial vehicle based on OCR recognition. Through OCR recognition technology, intelligent shooting and recognition methods such as fixed-point hovering of unmanned aerial vehicles, it provides great convenience for the acquisition of images of the light and dark lines acquisition module of the outdoor LED display screen, and reduces the risk and accuracy of high-altitude correction of the display screen; each module of the present invention is independently decoupled, which has the beneficial effect of facilitating the further expansion and development of a single module according to subsequent actual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a structural schematic diagram of a system for calibrating bright and dark lines of an LED display screen on a drone according to an embodiment of the present application;

[0053] Figure 2 This is a schematic diagram of the structure of an OCR recognition visual perception unit according to an embodiment of the present application;

[0054] Figure 3 This is a flow chart of a method for calibrating bright and dark lines of an LED display screen by a drone according to an embodiment of the present application;

[0055] Figure 4 The method of determining the starting point of the UAV flight route by OCR recognition of the visual perception unit in one embodiment of the present application; sequentially identifying the label at the center of each module of the LED display screen includes a step flow chart;

[0056] Figure 5 A flowchart of the steps of planning a flight route of a drone according to an embodiment of the present application;

[0057] Figure 6 A flowchart of the steps of outputting corresponding bright and dark line correction coefficients for the high-definition images corresponding to each module in one embodiment of the present application, and then directly feeding back the bright and dark line correction coefficients to the display screen bright and dark line correction unit to implement bright and dark line correction of each module;

[0058] Figure 7 A structural diagram of an LED display screen corresponding to a route planning module of an embodiment of the present invention, which is composed of modules No. 1-25;

[0059] Figure 8 This is a structural diagram of an embodiment of the present invention in which a plurality of lamp points are distributed around the center line of the joint seam between modules;

[0060] Fig. 9 This is a schematic diagram of the captured image of the upper left corner module 1 of the LED display screen and the next direction of the drone according to one embodiment of the present invention;

[0061] Fig.10 The present invention is applied for an embodiment of the present invention, which is a schematic diagram of the image collected by the middle module 2 at the top of the LED display screen and the next direction of the drone;

[0062] Fig.11 An embodiment of the present invention is a schematic diagram of the captured image of the upper right corner module 8 of the LED display screen and the next direction of the drone. DETAILED DESCRIPTION

[0063] In order to more clearly understand the above-mentioned objectives, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0065] like Figure 1 As shown, an embodiment of the present application discloses a drone correction LED display screen bright and dark line system 100, the system includes a drone control subsystem 200 and a display screen control subsystem 300; the drone control subsystem 200 includes: an OCR recognition visual perception unit 120, a route planning module 130, and a module image acquisition and transmission unit 110; the drone control subsystem 200 is arranged in a drone controller;

[0066] The display screen control subsystem 300 includes: a display screen bright and dark line correction unit 310; the display screen control subsystem 300 is arranged in the display screen controller;

[0067] In some optional embodiments, the display screen control subsystem 300 and the drone control subsystem 200 are positioned by the Beidou satellite positioning system or the GPS satellite positioning system; the display screen control subsystem 300 and the drone control subsystem are data transmitted by indoor or outdoor wireless communication; the wireless communication includes 4G\5G signal transmission;

[0068] like Figure 2As shown, the OCR recognition visual perception unit 120 includes a forward vision detection module 122, a forward vision OCR recognition module 121, a lower vision and ultrasonic sensor module 123, and a distance measurement perception module 124;

[0069] The forward vision detection module 122 is used to detect the LED display screen directly in front of the drone and select a horizontal fixed-point hovering position;

[0070] The lower visual and ultrasonic sensor module 123 is used to determine the longitudinal relative position of the drone to the ground;

[0071] The forward vision OCR recognition module 121 is used to determine the starting point of the UAV flight route through the OCR recognition visual perception unit, and sequentially identify the label at the center of each module of the LED display screen; specifically, the label can be a number, an image, a text or a combination thereof;

[0072] The route planning module 130 is used to plan the flight route of the drone, and the starting point of the drone flight route is used as the origin of the drone flight, and a fixed horizontal movement distance and a vertical movement distance are set according to different models of LED display screens; specifically, Figure 7 As shown, the module labels 1-25 identified in sequence by the front vision OCR recognition module are photographed one by one in sequence, and the steps of forming a photographing path include: taking the leftmost module 1 of the highest row of modules 1-8 of the LED display screen as the flight origin of the drone, flying rightward along the highest row of modules to the rightmost module 8 in sequence, then moving down one row, flying leftward from the rightmost module 9 to the leftmost module 16 in sequence, then moving down one row, and repeating the above steps; specifically, the flight route of the drone when taking photos is planned according to the shape of a "bow".

[0073] like Figure 7 As shown, for example, the LED display screen corresponding to the route planning module of the present invention is composed of modules No. 1-25, each module has a corresponding unique digital label, the black solid line between the modules is the joint between the modules, and the line with an arrow on the digital label is a route planning path; Figure 8As shown, taking the 15th module as an example, when the drone completes taking pictures from module 1 to module 3, the correction coefficients of the bottom 91 of module 1, the bottom 92 of module 2, and the bottom 93 of module 3 are obtained in sequence, that is, the corresponding correction coefficients of the top of module 16, the top of module 15, and the top of module 14, and so on; the top of module 15 and the bottom 92 of module 2 become common edges; all common edges of the LED display screen module only need to calculate the correction coefficient once; through such a route planning method and its calculation method, it has the beneficial effect of greatly reducing the amount of calculation for obtaining the correction coefficient of each module, making the system run more smoothly and the adjustment of light and dark lines faster and more accurate.

[0074] In some optional embodiments, the step of the drone flying along the flight route includes: taking the leftmost module of the highest row of modules of the LED display screen as the drone flight origin, flying rightward along the highest row of modules to the rightmost module, then moving down at least two rows, flying leftward from the rightmost module to the leftmost module, then moving down at least two rows, and repeating the above steps; specifically, planning the flight route of the drone when taking pictures according to the shape of a "bow"; at this time, the drone can take a high-definition image including at least two modules for use in identifying and adjusting bright and dark lines;

[0075] The module image acquisition and transmission unit 110 is used to take photos of the modules one by one in sequence according to the route planning and the shooting sequence, and the module labels sequentially recognized by the front vision OCR recognition module to form a shooting path, and then the module high-definition images obtained by the sequential shooting are transmitted to the display control subsystem in real time by the drone; specifically, the flight route of the drone when taking photos is planned according to the "bow" shape, the shooting sequence is based on the numbers recognized from 1 to 1000 by the front vision OCR recognition module, and the shooting path presents a "bow" shape, and then the real-time high-definition image transmission is performed through the OCUSync image transmission function, and the drone transmits it to the computer;

[0076] The module image acquisition and transmission unit 110 includes an acquisition module, and the acquisition module is used to acquire the specified module image, and the acquisition field of view of the camera on the drone will include the module image adjacent to the four sides of the specified module, combined with the specified route planning and OCR recognition technology, to complete the flight mission of acquiring images;

[0077] The ranging sensing module 124, which is used for the drone, will obtain the minimum safe sensing distance;

[0078] The display screen light and dark line correction unit 310 is used to output the corresponding light and dark line correction coefficient for the high-definition image corresponding to each module, and then directly feed back the light and dark line correction coefficient to the display screen light and dark line correction unit to achieve light and dark line correction for each module; for example, according to the Windows SDK function supported by the drone, the corresponding light and dark line correction coefficient is output for the high-definition image corresponding to each module to achieve light and dark line correction for each module.

[0079] See also Figure 3 , an embodiment of the present invention also provides a method for calibrating bright and dark lines of an LED display screen by a drone, the method comprising:

[0080] Step S900, detecting the LED display screen in front of the drone visually, and selecting a horizontal fixed-point hovering position; specifically, the fixed-point hovering position should be safe and reasonable;

[0081] Step S902: determining the longitudinal relative position between the UAV and the ground;

[0082] Step S904, planning the flight route of the drone, using the starting point of the drone flight route as the origin of the drone flight, and setting a fixed horizontal moving distance and a vertical moving distance according to different models of LED display screens;

[0083] Step S906, determine the starting point of the UAV flight route through the OCR recognition visual perception unit; identify the label at the center of each module of the LED display screen in turn; and take pictures of the modules one by one according to the route planning and the order of taking pictures, and form a photo path, and then transmit the high-definition images of the modules obtained by taking pictures in turn to the display screen control subsystem in real time by the UAV;

[0084] Step S908: For the high-definition image corresponding to each module, the corresponding bright and dark line correction coefficient is output, and then the bright and dark line correction coefficient is directly fed back to the display screen bright and dark line correction unit to achieve bright and dark line correction of each module.

[0085] See also Figure 4 In some optional embodiments, the method of determining the starting point of the UAV flight route by OCR recognition of the visual perception unit; sequentially identifying the label at the center of each module of the LED display screen sequentially identifying the label at the center of each module of the LED display screen includes the steps of:

[0086] Step S100, obtaining a labeled image data set of LED display screen modules, selecting the LED display screen in test mode, taking a photo of each module by a mobile phone, camera or other shooting equipment, and obtaining label information of the four sides and the center of a single module;

[0087] Step S102, acquiring a characteristic and diverse image data set by switching the light point interval and the brightness value of the light point; for example, the acquired characteristic and diverse image data set includes a total of 10,000 images, and the labels corresponding to each module have 10 different styles;

[0088] Step S104: Obtain an OCR recognition model and select an OCR text recognition model combination; for example, in the case of digital labels, select a CTPN+CRNN model combination with better OCR text recognition performance;

[0089] Step S106: retrain to obtain a corresponding model;

[0090] Specifically, the digital label of the LED module is composed of discrete light points, so it is necessary to retrain to obtain the corresponding model;

[0091] For example, training is performed under the deep learning framework Tensorflow-1.14.0+CUDA10.0+CuDNN7.6+Python3.6+Windows10 to obtain the Pb format storage model corresponding to CTPN+CRNN, which is the OCR recognition model;

[0092] Step S108, transplanting the OCR recognition model to the OCR recognition visual perception unit of the drone; for example, by implementing the offline call of the OCR recognition model in the programming language Java in the Android Studio-4.1.2 environment, embedding it as a digital recognition module into the visual perception system of the drone.

[0093] See also Figure 5 In some optional embodiments, the step of planning the flight route of the drone includes:

[0094] Step S200, determine the coordinates of the return position, and connect the drone handle via a mobile phone; specifically, open the DJI Fly program on the mobile phone, select the specified model of drone, connect the drone to the handle, and the signal recognition function will automatically refresh. When the number reaches 8 or more, the coordinates of the departure point can be accurately obtained, and the drone can take off normally through voice broadcast prompts, and the coordinates of the departure point and / or return point are determined;

[0095] Step S202, obtaining the starting coordinates of the flight route, and selecting the normal flight mode on the handle to make the drone quickly approach the LED display screen; specifically, the distance measurement perception module of the drone will obtain the minimum safe perception distance; for example, the minimum safe perception distance of DJI's Mavic AIR2 series is 0.5 meters, and at the same time, the camera angle of the drone is adjusted to 0°, that is, the angle perpendicular to the LED display screen, and the position of the drone is manually fine-tuned to quickly find the digital label corresponding to module 1 and determine the starting coordinates of the flight route;

[0096] Step S204, perform route planning and construct a route model of the flight route; specifically, route planning is a fixed route module embedded in the function of the drone's pointing flight; for each different type of display screen, the length and width of a single module will also be different, and modeling is performed according to different models of outdoor LED display screens. Based on the starting coordinates of the flight route, the drone is switched from normal flight mode to low-speed flight mode, and the horizontal distance of the lateral movement and the vertical distance of the longitudinal movement of the drone are set; for example, a route model of a "bow"-shaped flight route is constructed, and after the flight is completed according to the "bow"-shaped flight route planned, the drone returns to the return point by pressing a return button.

[0097] See also Figure 6 In some optional embodiments, the high-definition image corresponding to each module outputs a corresponding bright and dark line correction coefficient, and then directly feeds back the bright and dark line correction coefficient to the display screen bright and dark line correction unit to achieve bright and dark line correction of each module, including the steps of;

[0098] Step S300, calculate the correction coefficient of the light and dark lines of each module; specifically, first lock the flight starting point position through the OCR recognition visual perception unit, then take photos of the modules according to the planned flight route and the order of the recognized module numbers, and transmit the photos taken in real time based on the OCUSync image transmission function of the drone to the A-Deline mobile phone program to obtain the light and dark line correction coefficient of each module with a digital label and save it;

[0099] Step S302, compensation feedback correction coefficient; specifically, according to the order of module labels, when taking a picture of the current module, the compensation coefficient of the previous module is fed back to the LED display screen, and the cycle is repeated until the compensation of the correction coefficient of the last module is completed.

[0100] The present invention is a method for correcting the light and dark lines of an LED display screen by a drone based on OCR recognition. Through OCR recognition technology, intelligent shooting and recognition methods such as fixed-point hovering of drones, it provides great convenience for the image acquisition module of the light and dark lines of an outdoor LED display screen, reduces the risk of high-altitude correction of the display screen and the accuracy of correction; each module is independently decoupled, which is convenient for further expansion and development of a single module according to subsequent actual needs;

[0101] In some optional embodiments, the method for calibrating bright and dark lines of an LED display screen by a drone further includes:

[0102] Obtain the minimum safe perception distance between the drone and the LED display in real time;

[0103] If the safety sensing distance is less than the set value, an alarm is issued, and the drone is automatically made to fly parallel to the LED display screen at the minimum set value.

[0104] In some optional embodiments, the step of detecting the LED display screen directly in front of the drone visually further includes:

[0105] Set the joint seam between modules as the center line, and select two rows of lamp beads on both sides of the center line respectively;

[0106] Set any cross intersection between four adjacent modules as a non-lit light point.

[0107] In some optional embodiments, the module tags identified in sequence by the front vision OCR recognition module are photographed one by one in sequence to form a photographing path, which includes:

[0108] Take the leftmost module of the highest row of modules of the LED display screen as the flight origin of the drone, fly rightward along the highest row of modules to the rightmost module, then move down one row, fly leftward from the rightmost module to the leftmost module, then move down one row, and repeat the above steps.

[0109] In some optional embodiments, the common edge formed by each module of the LED display screen connected with other modules is calculated only once for the correction coefficient of the bright and dark lines of each module;

[0110] In some optional embodiments, the method of obtaining a labeled image data set of an LED display screen module, selecting the LED display screen in a test mode, taking a photo of each module by a mobile phone, a camera or other photographing device, and obtaining label information of four sides and the center of a single module, further includes the steps of:

[0111] When collecting images of a specified module, the camera on the drone will include the images of the modules adjacent to the four sides of the specified module, and complete the flight mission of collecting images by combining the specified route planning and OCR recognition technology. For example, in the case of digital labels, the flight mission of collecting images is completed accurately in a pattern of increasing the number by 1 from small to large. The specific image collection method is as follows: Figure 8-11 As shown, Fig. 9 This is the captured image of the upper left corner module 1 of the LED display screen, and the dotted arrow is the next direction of the drone; Fig.10 This is the captured image of the top of the LED display near the middle module 2, and the dotted arrow is the next direction of the drone; Fig.11 It is the collected image of the top right corner module 8 of the LED display screen, and the dotted arrow is the next direction of the drone; Figure 8 The image of module 15 in the middle of the LED display screen is collected. The drone then moves from module 15 to module 16 along the center line of module 15 and module 16.

[0112] like Figure 8 As shown, in some optional embodiments, when the UAV is preparing to take off according to the route plan, the present invention applies to adjust the LED display screen to the test mode. Taking module No. 15 as an example, when taking a photo of module No. 15, the UAV collects the module light point distribution map; through the collected module light point distribution map, it can be seen that there are many light points distributed on the LED display screen module; with the joint seam between modules as the center line, only at least two columns of lamp beads on both sides of the center line of many light points are selected to obtain and calculate high-definition images; it has the beneficial effect of reducing the image size and facilitating the subsequent massive calculation of correction coefficients; in some optional embodiments, the cross intersections of adjacent modules (for example, the cross intersection 94 of adjacent modules) do not light up the lights in order to facilitate the segmentation of each edge of a single module. In this way, the accuracy of the correction coefficient is higher.

[0113] The present invention application also provides a device for correcting bright and dark lines of an LED display screen by using a drone, and the device comprises: the method for correcting bright and dark lines of an LED display screen by using a drone as described in any embodiment of the present invention application.

[0114] The present invention application also provides a storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, implements a method for correcting bright and dark lines of an LED display screen using a drone as described in any embodiment of the present invention application.

[0115] Any embodiment of the present invention has the beneficial effect of improving the efficiency and accuracy of LED display screen calibration.

[0116] In the several specific embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system implementation described above is only illustrative, for example, the division of the components is only a logical function division, and there may be other division methods in actual implementation.

[0117] In addition, each functional module / component in each embodiment of the present invention may be integrated into the same processing module / component, or each module / component may exist physically separately, or two or more modules / components may be integrated into the same module / component. The above-mentioned integrated modules / components may be implemented in the form of hardware or in the form of hardware plus software functional modules / components.

[0118] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0119] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A drone LED display screen light and dark line correction system, characterized in that: The system includes a drone control subsystem and a display screen control subsystem; The drone control subsystem includes: an OCR recognition visual perception unit, a route planning module, and a module image acquisition and transmission unit; the drone control subsystem is arranged in a drone controller; The display screen control subsystem comprises: a display screen bright and dark line correction unit; the display screen control subsystem is arranged in a display screen controller; The OCR recognition visual perception unit includes a forward vision detection module, a forward vision OCR recognition module, a downward vision and ultrasonic sensor module, and a distance measurement perception module; The forward vision detection module is used to detect the LED display screen directly in front of the drone and can select a horizontal fixed-point hovering position; The lower visual and ultrasonic sensor modules are used to determine the longitudinal relative position of the drone to the ground; The forward vision OCR recognition module is used to determine the starting point of the UAV flight route through the OCR recognition visual perception unit, and sequentially identify the label in the center of each module of the LED display screen; The route planning module is used to plan the flight route of the drone, using the starting point of the drone flight route as the origin of the drone flight, and setting a fixed horizontal moving distance and a vertical moving distance according to different models of LED display screens; The module image acquisition and transmission unit is used to take photos of the modules one by one according to the route planning and the order of taking photos, and the module tags identified in sequence by the forward vision OCR recognition module to form a photo path, and then transmit the module high-definition images obtained by taking photos in sequence to the display control subsystem in real time by the drone; The module image acquisition and transmission unit includes an acquisition module; the acquisition module is used to acquire the specified module image, and the acquisition field of view of the camera on the drone will include the module image adjacent to the four sides of the specified module, combined with the specified route planning and OCR recognition technology, to complete the flight mission of acquiring images; The distance sensing module used in the drone will obtain the minimum safe sensing distance between the drone and the LED display screen; The display screen bright and dark line correction unit is used to output corresponding bright and dark line correction coefficients for the high-definition image corresponding to each module, and then directly feed back the bright and dark line correction coefficients to the display screen bright and dark line correction unit to achieve bright and dark line correction of each module.

2. A method for calibrating bright and dark lines of LED display screen by using drone, characterized in that: The method comprises: Detect the LED display screen in front of the drone's vision and select a horizontal fixed-point hovering position; Determine the longitudinal relative position of the drone to the ground; Plan the flight route of the drone, use the starting point of the drone flight route as the origin of the drone flight, and set a fixed horizontal moving distance and vertical moving distance according to different models of LED display screens; The starting point of the UAV flight route is determined by the OCR recognition visual perception unit; the label at the center of each module of the LED display screen is identified in turn; and according to the route planning and the order of taking photos, the module labels are sequentially identified by the forward vision OCR recognition module to take photos of the modules one by one to form a photo path, and then the high-definition images of the modules obtained by the sequential photos are transmitted by the UAV to the display control subsystem in real time; For each module's corresponding high-definition image, the corresponding light and dark line correction coefficient is output, and then the light and dark line correction coefficient is directly fed back to the display screen's light and dark line correction unit to achieve light and dark line correction for each module.

3. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The method of determining the starting point of the UAV flight route by using the OCR recognition visual perception unit and sequentially identifying the label at the center of each module of the LED display screen includes the following steps: Obtain a labeled image dataset of LED display modules, select the LED display to test mode, and take photos of each module using a mobile phone, camera, or other shooting equipment. The requirements for taking photos require obtaining label information of the four sides of a single module and the center of the module. By switching the interval between light points and the brightness value of light points, a characteristic and diverse image data set is obtained; the characteristic and diverse image data set obtained includes a total of 10,000 images, and the labels corresponding to each module have 10 different styles; Obtain the OCR recognition model and select the OCR text recognition model combination; select the CTPN+CRNN model combination with good OCR text recognition performance; Retrain to obtain the corresponding model; Transplant the OCR recognition model to the drone's OCR recognition visual perception unit.

4. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The steps of planning the flight route of the drone include: Determine the coordinates of the return location and connect the drone handle via the mobile phone; Get the starting coordinates of the flight route, and select the normal flight mode on the handle to make the drone quickly approach the LED display; Carry out route planning and construct a route model for the flight route.

5. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The corresponding bright and dark line correction coefficient is output for the high-definition image corresponding to each module, and then the bright and dark line correction coefficient is directly fed back to the bright and dark line correction unit of the display screen to realize the bright and dark line correction of each module, including the steps of: Calculate the correction coefficient of bright and dark lines of each module; Compensation feedback correction factor; Specifically, according to the order of the module labels, when taking a picture of the current module, the compensation coefficient of the previous module is fed back to the LED display screen, and the cycle is repeated until the compensation of the correction coefficient of the last module is completed.

6. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The method further comprises: Obtain the minimum safe perception distance between the drone and the LED display in real time; If the safety sensing distance is less than the set value, an alarm is issued, and the drone is automatically made to fly parallel to the LED display screen at the set value.

7. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The step of detecting the LED display screen directly in front of the drone vision also includes the following steps: Set the joint seam between modules as the center line, and select two rows of lamp beads on both sides of the center line respectively; Set any cross intersection between four adjacent modules as a non-lit light point.

8. The method for correcting bright and dark lines of LED display screen by using drone according to claim 2, characterized in that: The module tags identified in sequence by the forward vision OCR recognition module take photos of the modules one by one in sequence to form a photo path, including: Take the leftmost module of the highest row of modules of the LED display screen as the flight origin of the drone, fly rightward along the highest row of modules to the rightmost module, then move down one row, fly leftward from the rightmost module to the leftmost module, then move down one row, and repeat the above steps.

9. The method for correcting bright and dark lines of LED display screen by using drone according to claim 3, characterized in that: The method of obtaining a labeled image data set of an LED display screen module, selecting the LED display screen in a test mode, and photographing each module by a mobile phone, a camera, or other photographing equipment, wherein the photographing requirement requires obtaining label information of four sides of a single module and the center of the module, further comprises the following steps: When collecting images of a specified module, the camera's field of view on the drone will include module images adjacent to the four sides of the specified module, and combined with the specified route planning and OCR recognition technology, the flight mission of collecting images is completed.

10. A method for correcting bright and dark lines of LED display screen by using drone according to any one of claims 2-9, characterized in that: The correction coefficient of the bright and dark lines of each module is calculated only once for the common edge formed by each module of the LED display screen connected with other modules.

11. A device for calibrating bright and dark lines of LED display screen by drone, characterized in that: include: The method for calibrating bright and dark lines of an LED display screen using a drone as described in any one of claims 2-9.

12. A storage medium, characterized in that: A computer program is stored thereon, wherein the computer program, when executed by a processor, implements a method for correcting bright and dark lines of an LED display screen using a drone as claimed in any one of claims 2 to 9.

Images