Image processing method and system for LED special-shaped curved screen

Abstract

The present invention discloses an image processing method and system for LED special-shaped curved screens, the method comprising: S100: assembling and pre-debugging LED special-shaped curved screens, so that the LED special-shaped curved screens can display logically normal images; S101: establishing the light point spatial coordinates of all light points of the corresponding LED special-shaped curved screens by identifying each LED lamp bead in the LED special-shaped curved screen; S102: converting the light point spatial coordinates into spatial image pixel coordinates according to the resolution and field angle information of the LED special-shaped curved screen; S103: generating a preset resolution image that can completely cover the entire LED special-shaped curved screen, and then calculating the corresponding plane image according to the spatial image pixel coordinates; S104: rearranging the plane image again according to the light point spatial coordinates, and performing point-to-point driving, and finally completing the video display of the LED special-shaped curved screen. The present invention overcomes the problem of discontinuous pixel-level video images in LED special-shaped curved screens such as spheres, hemispherical screens or other hyperbolic surfaces.

Description

Technical Field

[0001] The present invention belongs to the technical field of LED display, and in particular relates to an image processing method and system for an LED special-shaped curved screen. Background Art

[0002] LED display screens are designed with semiconductor light-emitting diodes as the main components. Due to their advantages of high brightness, high contrast, long life and customizable display size, they have been widely used in outdoor commercial advertising and display, building curtain walls, airport station traffic guidance, indoor conference display, dispatch command display, school hospital information broadcast and other scenarios. With the continuous advancement and development of LED display technology, the demand for special application scenarios that require small pitch, high density, high refresh rate and high reliability is becoming more and more widespread. For example, in the visual display system in the field of simulation training, the existing immersive ball screen or ball belt screen visual display system is mostly realized by projection display technology, and the display function of the ball screen or ball belt screen is realized by multiple projectors in the form of image splicing and fusion. Projection technology belongs to passive display technology. In the visual display system, there are generally problems such as low display brightness, low contrast, low channel image pixel utilization, complex image processing, inability to achieve full field of view due to light path occlusion, short life and difficult maintenance. In order to improve and solve these problems, a visual display solution based on LED display technology is proposed. Because LED display screens have the advantages of high brightness, high contrast, wide color gamut, high refresh rate and full field of view, they can present better immersion and realism than projection technology in the visual display process of simulated training. In addition, their long life, easy installation and easy maintenance make LED as a visual display system have great research significance and application value.

[0003] The visual display system used in simulation training is usually designed in a full wrap, half wrap or partial wrap manner, with the purpose of providing trainees with a more realistic scene display and thus achieving better training results. Therefore, such visual display systems are usually designed into ball screens, ball belt screens, dome-like and other hyperbolic shapes. This curved deformed screen is called an LED special-shaped curved screen, which mainly uses semiconductor light-emitting diodes as basic display elements, and designs and processes LED light panels of different shapes, and splices them into the required shape. Then, according to the shape of the display screen, special processing is performed on single-channel or multi-channel video images to be displayed, and the LED special-shaped curved screen drive is completed, so that it can finally present a normal and continuous video picture. A display system. Generally, LED special-shaped curved screens are accompanied by two-dimensional bending deformation and LED point arrangement. Therefore, the corresponding image processing and display driving need to comprehensively consider the video image source, the shape of the LED special-shaped curved screen, the dot pitch, the resolution, the LED light board point arrangement and the hardware resources for image processing and driving. In the simulation training field, the display image clarity and the real-time performance of image processing and transmission must also be considered. This also makes the image processing methods, devices and systems of LED special-shaped curved screens the focus of research for people in this field.

[0004] In the field of simulation training with higher requirements, the visual display system is usually a ball screen or a ball strip screen. In the LED-based ball screen or ball strip screen, in order to ensure that the spherical screen can display the visual scene normally, the traditional processing method is: the visual generation system first generates a corresponding spherical image according to the field of view of the screen, and then uses the camera model or projection model to generate and output multiple plane graphics according to the required resolution. At this time, the plane graphic is a geometrically deformed image. If this image is displayed directly on the ball screen, more serious deformation will occur. In order to eliminate this deformation, a set of geometric correction modules will be added to the system. The main function is to perform an inverse transformation on the deformed image output by the front end, that is, to perform another nonlinear distortion correction, and finally obtain a video image that can be correctly displayed on the ball screen. Patent CN202011438776.3 (A spherical video image correction method) is a method for solving the geometric deformation of video images proposed based on the working principle of the visual display system in the field of simulation training, characterized by: a. Determine the display field angle parameters of each channel; b. Generate multi-channel rectangular graphics and synchronously output them to the back-end graphic corrector for geometric correction of the graphics from plane to sphere; c. The graphic corrector converts each channel image into a curved image; d. The graphic correction parameters are saved in FLASH; e. The projector model projects the pre-deformed image onto a hemispherical projection screen to present the real target image. The invention can realize the geometric transformation from a plane image to a spherical image by loading the parameters on the hardware, without the need for image acquisition, correction grid and feature point acquisition, and calculation of a limited display area.

[0005] Through actual tests, it is found that after applying the above method, the correct video image can be observed on the spherical screen, which solves the problem of abnormal display image caused by geometric deformation. However, there is also a phenomenon of pixel-level video image discontinuity in some display areas, among which the equatorial area has the best display effect, and this phenomenon is more obvious towards the pole. The reason for the analysis is: the geometric correction method is an operation based on mathematical formulas, and the correction process is accompanied by image scaling processing, that is, there is a loss of image pixels. Because all operations are based on mathematical formulas, the loss of image pixels is uncertain; at the same time, in order to realize the splicing of LED spherical screens, the LED small light boards are similar to trapezoids, so there is a phenomenon of LED lamp bead arrangement point extraction, that is, the closer to the short side of the trapezoid, the fewer the number of LED lamp beads are arranged. When the lost pixel points after geometric correction and scaling algorithm processing cannot match the discarded lamp beads in the LED light board arrangement, it is possible that the image pixels that should have been displayed cannot be displayed because there are no LED lamp beads.

[0006] At the same time, in the actual test process, it was found that when the LED ball screen displays dynamic video images, there will be discontinuous image transition in some areas, especially when observing image content with complex backgrounds. This phenomenon will be more obvious. And this phenomenon will exist inside each LED light board, rather than at the joint of the entire image channel. When the same moving image passes through each LED light board, the same LED light board will lose the same image pixels, and finally give the observer a feeling that the displayed video image flows like "water waves".

[0007] The above problems, on the one hand, will lead to the loss of some important information, and on the other hand, will give users who use the LED ball screen for a long time a feeling of dizziness, greatly reducing the performance of the LED ball screen in visual display. Therefore, how to effectively solve the pixel-level image loss phenomenon of LED special-shaped curved screen caused by image processing such as image geometric correction or scaling and LED light board point arrangement, which in turn leads to discontinuous transition of the displayed video image and the "water wave" display phenomenon is a problem that people in this field need to solve urgently. Summary of the invention

[0008] The purpose of the present invention is to overcome the problem of discontinuous pixel-level video images on LED special-shaped curved screens such as spheres, hemispherical screens or other hyperbolic surfaces, and disclose an image processing method and system for LED special-shaped curved screens.

[0009] On the one hand, the purpose of the present invention is achieved by the following technical solutions:

[0010] An image processing method for an LED special-shaped curved screen, the image processing method for an LED special-shaped curved screen comprising:

[0011] S100: Assemble and pre-debug the LED special-shaped curved screen so that the LED special-shaped curved screen can display logically normal images;

[0012] S101: by identifying each LED lamp bead in the LED special-shaped curved screen, the light point spatial coordinates of all the light points of the corresponding LED special-shaped curved screen are established;

[0013] S102: converting the light point spatial coordinates into spatial image pixel coordinates according to the resolution and field of view information of the LED special-shaped curved screen;

[0014] S103: Generate a preset resolution image that can completely cover the entire LED special-shaped curved screen, and then calculate the corresponding plane image according to the pixel coordinates of the spatial image;

[0015] S104: rearrange the plane image again according to the spatial coordinates of the light points, and perform point-to-point driving, and finally complete the video display of the LED special-shaped curved screen.

[0016] According to a preferred implementation paradigm, step S101 includes: by identifying each LED lamp bead in the LED light board, first establishing the plane coordinates of the light points in each LED light board in the LED special-shaped curved screen, and then integrating the plane coordinates of the light points into the spatial coordinates of the light points of the LED special-shaped curved screen according to the arrangement order of each LED light board.

[0017] According to a preferred implementation paradigm, step S102 includes:

[0018] Establish a spatial coordinate system, select the coordinates of the starting point, and represent a point in space with longitude and latitude;

[0019] According to the resolution and field of view parameters of the LED special-shaped curved screen, the longitude and latitude of the interval between adjacent LED lamp beads are calculated;

[0020] According to the row and column scanning driving principle of the video image, starting from the starting point of the light point spatial coordinates, taking the longitude and latitude interval as the stepping unit, traverse all the light point spatial coordinates in turn and increase the corresponding longitude and latitude coordinates;

[0021] Extract the longitude and latitude coordinates and add the corresponding channel number to get the spatial image pixel coordinates.

[0022] According to a preferred implementation paradigm, step S103 includes: firstly generating an image of a preset resolution that can completely cover the corresponding field of view angle according to the field of view angle parameters of the LED special-shaped curved screen to be displayed;

[0023] Then, the corresponding image pixels are intercepted from the generated preset resolution image according to the spatial image pixel coordinates, and the intercepted image pixels are rearranged in row and column order into a two-dimensional plane image with a fixed resolution. When there is no image pixel in a row, its value is filled with 0; finally, a plane image with a fixed resolution is output.

[0024] According to a preferred implementation paradigm, in step S103, the process of capturing an image of a preset resolution image is as follows:

[0025] The process of compressing the generated image of preset resolution to a corresponding size according to the image size determined by the pixel coordinates of the spatial image;

[0026] The image scaling algorithm used includes neighbor interpolation, bilinear interpolation, bicubic interpolation, cubic convolution interpolation or other algorithms.

[0027] On the other hand, the present invention also discloses:

[0028] An image processing system for an LED special-shaped curved screen, wherein the image processing system uses the above-mentioned image processing method to perform image processing on the LED special-shaped curved screen;

[0029] The image processing system includes a video image generation system, an image processing device, and an LED driving system;

[0030] The LED driving system is configured to establish the spatial coordinates of all the lamp points of the LED special-shaped curved screen, and perform point-to-point driving of the LED special-shaped curved screen, and finally complete the video display of the LED special-shaped curved screen;

[0031] The image processing device is configured to convert the light point spatial coordinates into spatial image pixel coordinates, and to rearrange the planar image;

[0032] The image generation system is configured to generate an image of a preset resolution that can completely cover the screen, calculate the corresponding plane image based on the spatial image pixel coordinates, and finally output it to the image processing device through a hardware interface.

[0033] According to a preferred implementation paradigm, the video image generation system outputs video images to an image processing device via a video interface; the image processing device outputs video images to an LED driving system via a video interface; the LED driving system is connected to the LED special-shaped curved screen via Ethernet and parallel data line interfaces.

[0034] According to a preferred implementation paradigm, the LED driving system includes a video sending card and a video receiving card, the video sending card and the video receiving card are interconnected by an Ethernet cable, one video sending card can be connected to several video receiving cards, and one video receiving card is connected to several LED light boards through a parallel data line;

[0035] The video receiving card identifies each LED lamp bead in each LED light board, first establishes the plane coordinates of the light point in each LED light board, and the video sending card receives the plane coordinates of the light point fed back by the video receiving card, and then integrates the plane coordinates of the light point into the spatial coordinates of the light point of the LED special-shaped curved screen according to the arrangement order of the LED light boards;

[0036] The plane coordinates of the light point include 7 parameters, 2 parameters representing the position information of the video receiving card, 2 parameters representing the position information of the LED light board connected to the corresponding video receiving card, 2 parameters representing the row and column coordinates of the light point in a certain LED light board, and 1 parameter representing a flag indicating whether the corresponding light point in the corresponding LED light board actually exists;

[0037] The light point spatial coordinates include 4 parameters, wherein 1 parameter represents the channel number of the LED driving system, 2 parameters represent the row and column coordinates of the light point in the LED special-shaped curved screen, and 1 parameter represents a flag indicating whether the corresponding light point actually exists.

[0038] According to a preferred implementation paradigm, the image processing device includes a computing unit, a storage unit, and a video image real-time processing unit;

[0039] The calculation unit is configured to convert the light point spatial coordinates obtained by the LED driving system into spatial image pixel coordinates, and feed back the obtained spatial image pixel coordinates to the front-end video image generation system;

[0040] The storage unit is configured to store the spatial coordinates of the light point acquired by the LED driving system;

[0041] The video image real-time processing unit is configured to process high-resolution and high-refresh-rate video images in real time.

[0042] According to a preferred implementation paradigm, the spatial image pixel coordinates include three parameters, wherein one parameter represents the video channel sequence number, and two parameters represent the image pixel coordinates that need to be obtained in the video image generation system.

[0043] The aforementioned main scheme of the present invention and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection by the present invention. After understanding the scheme of the present invention, those skilled in the art can understand that there are multiple combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present invention, and they are not exhaustively listed here.

[0044] Beneficial effects of the present invention:

[0045] The method and system of the present invention solve the problems of discontinuous image transition and "water waves" in video images in traditional methods by matching image compression loss points with the one-to-one correspondence of the light beads arranged on the special-shaped curved screen.

[0046] The system of the present invention does not need to perform geometric correction from a space image to a plane image, and does not need to perform image interpolation algorithm processing on hardware, thereby reducing the complexity of hardware processing and improving the real-time performance of the system.

[0047] All image processing algorithms of the method and system of the present invention are implemented in a high-performance computer (video image generation system). Because it can perform floating-point operations, an algorithm with better performance can be selected to perform interpolation algorithm processing on the image, thereby ensuring the continuity of the output image source. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic flow chart of the method of the present invention;

[0049] Figure 2 It is a schematic diagram of coordinate transformation of the system of the present invention;

[0050] Figure 3 It is a schematic diagram of coordinate transformation of the system of the present invention;

[0051] Figure 4 It is a schematic diagram of mapping the converted video image and the LED special-shaped curved screen of the present invention;

[0052] Figure 5 It is a schematic diagram of the interval between the latitude lines of two adjacent lamp beads;

[0053] Among them, 10-LED light board, 101-each LED lamp bead in the LED light board, 20-plane coordinates of LED light points, 201-effective light points driven, 202-invalid light points driven, 30-spatial coordinates of LED special-shaped curved screen light points, 301-effective light points driven, 302-invalid light points driven, 40-spatial image pixel coordinates, 401-effective pixels of spatial image, 50-plane image output by the video image generation system, 501-effective pixels of the plane image, 502-invalid pixels filled in the plane image, 60-rearranged plane image, 601-effective pixels to be displayed, 602-invalid pixels discarded. DETAILED DESCRIPTION

[0054] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0055] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. In addition, the present invention should point out that in the present invention, if the specific structure, connection relationship, position relationship, power source relationship, etc. are not specifically written, the structure, connection relationship, position relationship, power source relationship, etc. involved in the present invention are all known to those skilled in the art based on the existing technology without creative work.

[0056] On the one hand, the present invention discloses an image processing method for an LED special-shaped curved screen, referring to Figure 1 As shown, the image processing method of the LED special-shaped curved screen includes the following steps.

[0057] Step S100: Assemble and pre-debug the LED special-shaped curved screen so that the LED special-shaped curved screen can display a logically normal image.

[0058] Step S101: by identifying each LED lamp bead in the LED special-shaped curved screen, the lamp point spatial coordinates of all lamp points of the corresponding LED special-shaped curved screen are established.

[0059] Preferably, step S101 includes: by identifying each LED lamp bead in the LED light board, first establishing the plane coordinates of the light points in each LED light board in the LED special-shaped curved screen, and then integrating the plane coordinates of the light points into the spatial coordinates of the light points of the LED special-shaped curved screen according to the arrangement order of each LED light board.

[0060] Step S102: converting the light point spatial coordinates into spatial image pixel coordinates according to the resolution and field angle information of the LED special-shaped curved screen.

[0061] Preferably, step S102 includes:

[0062] Establish a spatial coordinate system, select the coordinates of the starting point, and represent a point in space with longitude and latitude;

[0063] According to the resolution and field of view parameters of the LED special-shaped curved screen, the longitude and latitude of the interval between adjacent LED lamp beads are calculated;

[0064] According to the row and column scanning driving principle of the video image, starting from the starting point of the light point spatial coordinates, taking the longitude and latitude interval as the stepping unit, traverse all the light point spatial coordinates in turn and increase the corresponding longitude and latitude coordinates;

[0065] Extract the longitude and latitude coordinates and add the corresponding channel number to get the spatial image pixel coordinates.

[0066] Step S103: Generate a preset resolution image that can completely cover the entire LED special-shaped curved screen, and then calculate the corresponding plane image according to the pixel coordinates of the spatial image.

[0067] Preferably, step S103 comprises: firstly generating an image of preset resolution that can completely cover the corresponding field of view angle according to the field of view angle parameters of the LED special-shaped curved screen to be displayed;

[0068] Then, the corresponding image pixels are intercepted from the generated preset resolution image according to the spatial image pixel coordinates, and the intercepted image pixels are rearranged in row and column order into a two-dimensional plane image with a fixed resolution. When there is no image pixel in a row, its value is filled with 0; finally, a plane image with a fixed resolution is output.

[0069] Furthermore, in step S103, the process of capturing an image of the preset resolution image is as follows:

[0070] The process of compressing the generated image of preset resolution to a corresponding size according to the image size determined by the pixel coordinates of the spatial image;

[0071] The image scaling algorithms used include neighbor interpolation, bilinear interpolation, bicubic interpolation, cubic convolution interpolation or method.

[0072] Step S104: rearrange the plane image again according to the spatial coordinates of the light points, and perform point-to-point driving, and finally complete the video display of the LED special-shaped curved screen.

[0073] Preferably, the rearrangement process is to insert invalid pixels at specific positions between valid pixels; since the invalid pixels in the image after rearrangement correspond one-to-one to the LED lamp beads extracted from the LED special-shaped curved screen, point-to-point driving of the image effective pixels and the LED physical lamp beads is achieved, eliminating the phenomenon of image pixel discontinuity caused by valid pixels being mistakenly discarded.

[0074] In another aspect, the present invention also discloses: an image processing system for LED special-shaped curved screen, wherein the image processing system uses the above-mentioned image processing method to process the image displayed on the LED special-shaped curved screen. The image processing system includes a video image generation system, an image processing device, and an LED driving system. Figure 2 shown.

[0075] Preferably, the video image generation system outputs video images to an image processing device via a video interface; the image processing device outputs video images to an LED driving system via a video interface; the LED driving system is connected to the LED special-shaped curved screen via Ethernet and parallel data line interfaces.

[0076] Furthermore, the video image generation system outputs the video image to the image processing device through a video interface such as DP, HDMI or VGA, and the image processing device outputs the video image to the LED driving system through a video interface such as DP, HDMI or VGA.

[0077] The LED driving system is configured to establish the spatial coordinates of all the lamp points of the LED special-shaped curved screen, and perform point-to-point driving of the LED special-shaped curved screen, and finally complete the video display of the LED special-shaped curved screen; the image processing device is configured to convert the spatial coordinates of the lamp points into spatial image pixel coordinates, and rearrange the plane image; the image generation system is configured to generate a preset resolution image that can completely cover the screen, and calculate the corresponding plane image based on the spatial image pixel coordinates, and finally output it to the image processing device through the hardware interface.

[0078] Preferably, the pre-debugging process of the LED special-shaped curved screen so that it can display logically normal image pictures includes: debugging the video image generation system to be able to output video image pictures normally, the image processing device to be able to receive video images and output them normally, the LED driving system to be able to drive each LED light board, and debugging to achieve the display order of each LED light board, and the entire LED special-shaped curved screen can display logically correct video images.

[0079] Preferably, the LED driving system includes a video sending card and a video receiving card, the video sending card and the video receiving card are interconnected by an Ethernet cable, one video sending card can be connected to several video receiving cards, and one video receiving card is connected to several LED light boards through parallel data lines.

[0080] Furthermore, establishing the spatial coordinates of all the lamp points of the LED special-shaped curved screen mainly includes two processes. First, the video receiving card identifies each LED lamp bead in each LED light board and first establishes the plane coordinates of the lamp points in each LED light board; second, the video sending card receives the plane coordinates of the lamp points fed back by the video receiving card, and then integrates the plane coordinates of the lamp points into the spatial coordinates of the lamp points of the LED special-shaped curved screen according to the arrangement order of the LED light boards.

[0081] The plane coordinates of the light point include 7 parameters, 2 parameters represent the position information of the video receiving card, 2 parameters represent the position information of the LED light board connected to the corresponding video receiving card, 2 parameters represent the row and column coordinates of the light point in a certain LED light board, and 1 parameter represents a flag indicating whether the corresponding light point in the corresponding LED light board actually exists; when there is a point sampling arrangement in the LED light board, the LED light points processed by the point sampling are indicated as not existing, and other physical LED light points that have not been processed by the point sampling are indicated as existing.

[0082] The light point spatial coordinates include 4 parameters, wherein 1 parameter represents the channel number of the LED driving system, 2 parameters represent the row and column coordinates of the light point in the LED special-shaped curved screen, and 1 parameter represents a flag indicating whether the corresponding light point actually exists.

[0083] Preferably, the image processing device comprises a computing unit, a storage unit and a video image real-time processing unit.

[0084] The calculation unit is configured to convert the light point spatial coordinates obtained by the LED driving system into spatial image pixel coordinates, and feed the obtained spatial image pixel coordinates back to the front-end video image generation system. The spatial image pixel coordinates include 3 parameters, 1 parameter represents the video channel sequence number, and 2 parameters represent the image pixel coordinates that need to be obtained in the video image generation system.

[0085] The storage unit is configured to store the spatial coordinates of the lamp points obtained by the LED driving system; the video image real-time processing unit mainly includes a video interface codec chip, an FPGA and a dynamic random access memory, etc., and is configured to process high-resolution and high-refresh-rate video images in real time.

[0086] Preferably, the video image generation system is composed of high-performance graphics computer hardware and software running on the computer hardware that can generate video images in real time.

[0087] Example

[0088] This embodiment discloses an image processing system for an LED special-shaped curved screen. The image processing system includes a video image generation system, an image processing device, and an LED driving system. The image processing system implements image processing according to the following steps.

[0089] Step S1: Complete the installation of the LED special-shaped curved screen and image processing system, including first splicing all the LED small light panels into an LED special-shaped curved screen according to the design serial number, and then connecting the video image generation system output to the image processing device input and the image processing device output to the LED drive system input through video cables such as DP, HDMI, DVI or VGA; the LED drive system includes a video sending card and a video receiving card, usually a video sending card is connected to multiple video receiving cards through an Ethernet cable, and a video receiving card is connected to multiple LED light panels through parallel data cables; finally, the power supply system is installed to ensure that all systems or equipment can be powered on and work normally.

[0090] After completing the system installation, start pre-debugging of the LED special-shaped curved screen, that is, first debug the LED drive system to ensure that the video sending card can recognize all video receiving cards, and then recognize all LED light boards; then adjust the control logic of the video sending card and the video receiving card from left to right and from top to bottom according to the image display principle, so that the LED special-shaped curved screen can display logically normal images; finally, start the video image generation system and image processing equipment, and debug so that the video image output by the video image generation system can be displayed on the LED special-shaped curved screen.

[0091] Since the LED special-shaped curved screen has the phenomenon of physical point extraction in the arrangement of LED lamp beads, and this problem is not taken into account during the pre-debugging process, the graphics displayed by the LED special-shaped curved screen will inevitably have problems such as image discontinuity and display misalignment.

[0092] Step S2: The LED special-shaped curved screen is made up of a variety of LED light boards. The LED light board 10 is approximately fan-shaped. At the same time, the arrangement of the LED lamp beads 101 inside the LED light board 10 has the characteristic of being arranged along the arc edge. If the LED lamp beads arranged in an arc are unfolded into a plane, it is found that the lamp beads on the same arc are arranged in a row, and the number of lamp beads on different arcs is not necessarily equal. It can be considered that different arcs correspond to different rows, and the number of lamp beads in each row is not necessarily equal. This unequal phenomenon is caused by the arrangement of points. Therefore, when designing the LED drive, it is also implemented by scanning point by point in the direction of the arc. Although lamp bead points appear in some rows, the drive is still driven with the maximum number of rows and the maximum number of columns. That is, the LED lamp beads that do not exist after being extracted will still retain their information in the drive system, but the drive system will perform corresponding discard processing when the drive scan reaches this point. Therefore, when acquiring and converting coordinates, the coordinates recognized by the LED drive system must also be used as a reference. The specific steps are:

[0093] a. The video receiving card in the LED drive system first identifies the LED light board it is connected to and records the arrangement position of the LED light board during the pre-debugging process (d x ,dy ); then identify each light point to be driven in the LED light board by controlling the scanning driver chip, and record the row and column coordinates and status information of the light point; finally establish the plane coordinates 20 (d x ,d y ,x,y,f), where (d x ,d y ) represents the arrangement position of the light board managed by a video receiving card, (x, y) represents the row and column coordinates of the light point to be driven in a single light board, and f represents the effective state of the light point. Figure 3 In the figure, 201 represents a valid lamp point driven, which can be recorded as a value of "1" or "T", and 202 represents an invalid lamp point driven, which can be recorded as a value of "0" or "F".

[0094] b. The video sending card in the LED drive system first identifies the video receiving card it is connected to and records the arrangement position of the video receiving card during the pre-debugging process (j x ,j y ); then read the plane coordinates of the light point stored in the video receiving card 20 (d x ,d y ,x,y,f); finally, combined with the arrangement position of the video receiving card (j x ,j y ) calculates the spatial coordinates of the light point in the entire screen 30 (t, x', y', f), where t represents the channel number of the driving video, that is, when the resolution and display field of the LED special-shaped curved screen are very large, the existing video image generation, processing and transmission system hardware cannot process the high-resolution video image, and the display drive of the entire screen is usually achieved by splicing video images of multiple channels. (x', y') represents the row and column coordinates of the light point to be driven in a single channel of the entire special-shaped curved screen. The specific calculation process is as follows:

[0095]

[0096] where x max Indicates the maximum number of lights in a row on a light board, y max Indicates the maximum number of lights in a row on a light board, all of which are positive integers; X max Indicates the maximum number of lights in a row composed of all light boards managed by a video receiving card, Y max Indicates the maximum number of light points in a row composed of all light boards managed by a video receiving card, all of which are positive integers; and there is d x ∈[0,n],j x ∈[0,m],d y ∈[0,h]],j y ∈[0,k],n,m,h,k are also all positive integers.

[0097] Figure 3 In the LED special-shaped curved screen lamp point spatial coordinates 30, f also represents the effective state of the lamp point, such as 301 represents a valid lamp point driven, which can be recorded as a value "1" or "T", and 302 represents an invalid lamp point driven, which can be recorded as a value "0" or "F".

[0098] c. The LED driving system transmits the calculated light point spatial coordinates 30 (t, x', y', f) of the LED special-shaped curved screen to the image processing device.

[0099] Step S3: The image processing device converts the spatial coordinates of the light point into spatial image pixel coordinates according to the resolution and field of view of the LED special-shaped curved screen, and sends the coordinate information to the front-end video image generation system.

[0100] The special-shaped curved screen is made up of a variety of LED light panels, which are characterized by getting smaller from the middle to the two ends. For ease of understanding, a sphere is used as an example. For a sphere, the distribution of LED lamp beads is usually arranged at equal intervals along the longitude and latitude of the sphere, with the largest number of lamp beads arranged on the equator and decreasing toward the poles. Since the meridian direction is arranged at equal intervals, there is no sampling point in the meridian direction, and the number of lamp beads is equal. At this time, first establish a spatial coordinate system with the center of the sphere as the origin, and agree that all lamp bead coordinates on the sphere are expressed in longitude and latitude. In order to map the spatial coordinates of the lamp point into spatial image pixel coordinates, perform the following operations:

[0101] a. First, the physical resolution W×H of the LED special-shaped curved screen can be calculated based on the spatial coordinates of the light point 30 (t, x', y', f), where The table shows the maximum value of the remaining lamp points on the latitude after removing the invalid lamp points from the lamp point spatial coordinates 30, that is, the maximum value of the lamp beads on the equator; The expression shows the maximum value of the remaining light points on the meridian after removing invalid light points from the light point spatial coordinates 30. The number can also be understood as the number of latitude layers on the LED special-shaped curved screen. According to the above characteristics of the LED special-shaped curved screen, W is a variable, that is, due to the existence of sampling points, the number of lamp beads at different latitudes may be different, and the number of latitude layers of the screen is fixed, so H can be considered as a fixed value.

[0102] b. Use measurement or design data to obtain the field of view parameters FOV of the LED curved screen X and FOV Y , where FOV X Indicates the horizontal field of view of the screen, and corresponds to W, FOV Y Indicates the vertical field of view, and corresponds to H. Then, according to the LED screen resolution and field of view parameters, the latitude and longitude intervals between two adjacent LED lamp beads are calculated, specifically:

[0103]

[0104] Among them, Δlon represents the interval between two adjacent lamp beads in latitude, Δlat represents the interval between two adjacent lamp beads in longitude. More specifically, Δlan actually represents the interval between the latitude lines of two adjacent lamp beads, such as Figure 5 shown.

[0105] c. Identify the starting point M(0,0) from the spatial coordinates 30(t,x',y',f) of the light point, and use this point as the starting point to identify the valid light points in turn according to the row and column scanning direction of the image; at the same time, assign spatial longitude and latitude coordinate values ​​to each identified valid light point, and the change of the longitude and latitude coordinate values ​​is in units of (Δlon, Δlan), so we can get the starting point coordinates M(0,0,0,0), and traverse in the row and column scanning direction of the image to get M1(1,0,Δlon,0), M2(2,0,2Δlon,0), M3(3,0,3Δlon,0), ..., P = (x',y',lon,lat), where point P represents any valid light point in the spatial coordinates and its corresponding spatial longitude and latitude coordinate values, which can be specifically expressed as:

[0106]

[0107] Wherein w and h are both positive integers.

[0108] d. The coordinates obtained through traversal are all the coordinates of the valid light points in the spatial coordinates 30 of the LED special-shaped curved screen light points. At this time, only the longitude and latitude coordinates of each point are retained (t, lon, lat). The coordinates expressed in the form of longitude and latitude are called spatial image pixel coordinates, and finally the coordinates are sent to the front-end video image generation system.

[0109] Step S4: The video image generation system generates an image. Specifically, the system first establishes a model with the same shape as the LED screen (mainly based on the displayed field of view parameters); then generates a high-resolution plane graphic, and further fits the plane graphic to the surface of the model to ensure that the plane graphic can completely cover the generated model; finally, according to the camera model, the plane graphic is obtained from the model and output through the hardware interface.

[0110] According to the above principle, the main process of this method is:

[0111] a. The video image generation system first calculates the field of view angle parameter FOV of the LED curved screen X and FOV Y Establish the same special-shaped surface model;

[0112] b. Then analyze the pixel coordinates of the spatial image sent by the image processing device to determine the maximum resolution of the spatial image (which is also the maximum resolution of the LED special-shaped curved screen), and further generate a high-resolution image that can completely cover the screen. For example, the maximum resolution of the LED special-shaped curved screen is 1920 (W) × 1080 (H). At this time, the video image generation system can generate 1920 (W) × 1080 (H), 3840 (W) × 2160 (H), 7680 (W) × 4320 (H) or any other high-resolution image greater than or equal to this resolution.

[0113] c. Further, the high-resolution image is overlaid on the generated model. Since the model is a hyperbolic surface, the transformation from a plane graphic to a curved surface must involve pixel compression. In this process:

[0114] 1) The pixel compression process uses floating-point arithmetic, allowing decimals to represent pixel positions in the image;

[0115] 2) After pixel compression, the number of pixels is also greater than or equal to the resolution of the LED special-shaped curved screen;

[0116] 3) With the help of high performance of computers, pixel compression can use better interpolation algorithms, including but not limited to neighboring interpolation, bilinear interpolation, bicubic interpolation and cubic convolution interpolation.

[0117] d. Finally, obtain the plane image from the model image and output it in the form of a hardware interface. The common method is to use a camera model, that is, place a camera at a certain position in the model, and then obtain the plane image by taking pictures with the camera. This method will involve the problem of later geometric correction of graphics. Therefore, in order to avoid this problem, this method is mainly as follows: first, analyze the number of pixels at each latitude from the spatial image pixel coordinates, and then determine the image pixels at the corresponding latitude in the generated image according to the latitude coordinates of these pixels; since the spatial image pixel coordinate values ​​generated by the image processing device are all integers, at this time, you can choose to directly read the pixel values ​​of the corresponding longitude and latitude coordinates from the model and save them, or you can further compress the pixels with decimal points in the model according to the actual number of pixels. The number of compressed pixels corresponds exactly to the number of pixels in the spatial image. The pixel compression is also mainly an interpolation algorithm, including but not limited to neighboring interpolation, bilinear interpolation, bicubic interpolation and cubic convolution interpolation. After comparison, the two methods of directly reading integer coordinate pixel values ​​and obtaining pixel values ​​by secondary compression have little difference in effect, so either one can be selected. Finally, the pixel values ​​intercepted from the model are arranged in sequence according to the row and column scanning direction of the image, and the positions without pixel values ​​in a row are all filled with "0" to obtain the plane image to be output.

[0118] Step S5: The image processing device performs real-time image processing on the video image by means of an internally designed FPGA+DDR SDRAM architecture.

[0119] When the image processing device receives a video image output from the video image generation system, it first caches the current frame plane image of the video image into the DDR SDRAM through the FPGA. At the same time, it reads the spatial coordinates of the LED special-shaped curved screen lamp points stored in the storage unit, and adjusts the pixel points in the plane image in sequence, that is, identifies the plane image valid pixels 501 in the plane image 50 output by the video image generation system, and then arranges them in sequence to the driven valid lamp points in the lamp point spatial coordinates. Finally, a new rearranged plane image 60 corresponding to the spatial coordinates of the LED special-shaped curved screen lamp points is obtained and stored in the DDR SDRAM. The FPGA then realizes the video image output by reading the plane image in the DDR SDRAM.

[0120] Since the rearranged plane image 60 received by the LED driving system corresponds one-to-one with the arrangement of the lamp points driven by the internal scanning, that is, the valid pixels in the image correspond to the valid lamp points to be driven, and the invalid pixels in the image can also correspond to the invalid lamp points to be driven, the LED driving system no longer cares about the content of the video image, and only needs to drive the LED lamp beads point-to-point. Finally, the invalid pixels correspond to the random lamp beads, and all the valid pixels correspond to the actual LED lamp beads, so as to achieve a continuous display effect without loss of the image.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An image processing method for LED special-shaped curved screen, characterized in that: The image processing method of the LED special-shaped curved screen includes: S100: Assemble and pre-debug the LED special-shaped curved screen so that the LED special-shaped curved screen can display logically normal images; S101: by identifying each LED lamp bead in the LED special-shaped curved screen, the light point spatial coordinates of all the light points of the corresponding LED special-shaped curved screen are established; Step S101 includes: by identifying each LED lamp bead in the LED lamp board, firstly establishing the plane coordinates of the lamp points in each LED lamp board in the LED special-shaped curved screen, and then integrating the plane coordinates of the lamp points into the spatial coordinates of the lamp points of the LED special-shaped curved screen according to the arrangement order of each LED lamp board; S102: converting the light point spatial coordinates into spatial image pixel coordinates according to the resolution and field of view information of the LED special-shaped curved screen; Step S102 includes: Establish a spatial coordinate system, select the coordinates of the starting point, and represent a point in space with longitude and latitude; According to the resolution and field of view parameters of the LED special-shaped curved screen, the longitude and latitude of the interval between adjacent LED lamp beads are calculated; According to the row and column scanning driving principle of the video image, starting from the starting point of the light point spatial coordinates, taking the longitude and latitude interval as the stepping unit, traverse all the light point spatial coordinates in turn and increase the corresponding longitude and latitude coordinates; Extract the longitude and latitude coordinates and add the corresponding channel number to get the spatial image pixel coordinates; S103: Generate a preset resolution image that can completely cover the entire LED special-shaped curved screen, and then calculate the corresponding plane image according to the pixel coordinates of the spatial image; Step S103 includes: firstly, generating a preset resolution image that can completely cover the corresponding field of view angle according to the field of view angle parameters of the LED special-shaped curved screen to be displayed; Then, the corresponding image pixels are intercepted from the generated preset resolution image according to the spatial image pixel coordinates, and the intercepted image pixels are rearranged in row and column order to form a two-dimensional plane image with a fixed resolution. When there is no image pixel in a row, its value is filled with 0; finally, the plane image with a fixed resolution is output; S104: rearrange the plane image again according to the spatial coordinates of the light points, and perform point-to-point driving, and finally complete the video display of the LED special-shaped curved screen.

2. The image processing method of the LED special-shaped curved screen according to claim 1, characterized in that: In step S103, the process of capturing an image of the preset resolution is as follows: The process of compressing the generated image of preset resolution to a corresponding size according to the image size determined by the pixel coordinates of the spatial image; The image scaling algorithms used include neighbor interpolation, bilinear interpolation, bicubic interpolation, and cubic convolution interpolation.

3. An image processing system for LED special-shaped curved screen, characterized in that: The image processing system uses the image processing method according to any one of claims 1 to 2 to perform image processing on the LED special-shaped curved screen; The image processing system includes a video image generation system, an image processing device, and an LED driving system; The LED driving system is configured to establish the spatial coordinates of all the lamp points of the LED special-shaped curved screen, and perform point-to-point driving of the LED special-shaped curved screen, and finally complete the video display of the LED special-shaped curved screen; The image processing device is configured to convert the light point spatial coordinates into spatial image pixel coordinates, and to rearrange the planar image; The image generation system is configured to generate an image of a preset resolution that can completely cover the screen, calculate the corresponding plane image based on the spatial image pixel coordinates, and finally output it to the image processing device through a hardware interface.

4. The image processing system according to claim 3, wherein: The video image generation system outputs the video image to the image processing device through the video interface; the image processing device outputs the video image to the LED driving system through the video interface; the LED driving system is connected to the LED special-shaped curved screen through Ethernet and parallel data line interfaces.

5. The image processing system according to claim 3, characterized in that: The LED driving system includes a video sending card and a video receiving card, the video sending card and the video receiving card are interconnected by an Ethernet cable, one video sending card can be connected to several video receiving cards, and one video receiving card is connected to several LED light boards through parallel data cables; The video receiving card identifies each LED lamp bead in each LED light board, first establishes the plane coordinates of the light point in each LED light board, and the video sending card receives the plane coordinates of the light point fed back by the video receiving card, and then integrates the plane coordinates of the light point into the spatial coordinates of the light point of the LED special-shaped curved screen according to the arrangement order of the LED light boards; The plane coordinates of the light point include 7 parameters, 2 parameters representing the position information of the video receiving card, 2 parameters representing the position information of the LED light board connected to the corresponding video receiving card, 2 parameters representing the row and column coordinates of the light point in a certain LED light board, and 1 parameter representing a flag indicating whether the corresponding light point in the corresponding LED light board actually exists; The light point spatial coordinates include 4 parameters, wherein 1 parameter represents the channel number of the LED driving system, 2 parameters represent the row and column coordinates of the light point in the LED special-shaped curved screen, and 1 parameter represents a flag indicating whether the corresponding light point actually exists.

6. The image processing system according to claim 3, characterized in that: The image processing device includes a computing unit, a storage unit and a video image real-time processing unit; The calculation unit is configured to convert the light point spatial coordinates obtained by the LED driving system into spatial image pixel coordinates, and feed back the obtained spatial image pixel coordinates to the front-end video image generation system; The storage unit is configured to store the spatial coordinates of the light point acquired by the LED driving system; The video image real-time processing unit is configured to process high-resolution and high-refresh-rate video images in real time.

7. The image processing system according to claim 3, characterized in that: The spatial image pixel coordinates include three parameters, wherein one parameter represents the video channel sequence number, and two parameters represent the image pixel coordinates that need to be obtained in the video image generation system.

Images