High-resolution image splicing television curtain wall composition method and system

Abstract

The invention relates to the technical field of image processing, in particular to a high-resolution image splicing television curtain wall composition method and system. The method comprises the following steps: acquiring a whole curtain wall display image; extracting boundary structure feature points at the spliced boundary; constructing a geometric deviation inversion model; estimating translation deviation along the splicing boundary direction; correcting the display unit image; evaluating the splicing geometric continuity based on the splicing boundary normal gradient consistency; carrying out iterative correction on the geometric deviation parameters; and outputting a final spliced display image. Compared with a splicing method depending on manual correction in the prior art, the technical problem that geometric dislocation of a splicing boundary is prone to occurring especially under the condition that a tiny rotation deflection angle exists in high-resolution and large-size television curtain wall installation is solved. Due to the fact that a geometric deviation correction mechanism and a closed-loop self-adaptive feedback correction mechanism based on normal gradient consistency are introduced, the geometric consistency of image splicing is improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method and system for assembling a high-resolution image splicing television wall. Background Technology

[0002] Currently, with the development of ultra-high-definition display technology, large-size, high-resolution video walls are widely used in command and dispatch, monitoring and display, digital exhibition halls, and professional image analysis scenarios. These video walls are typically composed of multiple display units spliced ​​together, with the content displayed in each unit being synchronously controlled to form the overall display image. Under high-resolution application conditions, the geometric continuity of the spliced ​​image directly affects the overall display effect and the accuracy of subsequent image analysis.

[0003] For image splicing in video walls, most existing technologies rely on manual calibration during installation or one-time geometric correction based on fixed parameters to reduce splicing errors by pre-adjusting the positional relationship of display units or statically transforming the displayed image. However, in practical engineering applications, the installation of display units inevitably involves factors such as slight rotation angles, uneven surfaces, and assembly errors. These errors are difficult to detect with the naked eye but are significantly amplified under high-resolution splicing display conditions, easily causing geometric misalignment and structural discontinuities at the splicing boundaries.

[0004] Furthermore, existing splicing correction methods typically lack effective means of evaluating the geometric consistency of the splicing results. The correction process is mostly open-loop control. Once the installation environment changes or the display unit undergoes a small displacement after long-term operation, the original correction parameters are difficult to correct in time, which can easily lead to a gradual decline in splicing quality and fail to fully meet the requirements for high-precision and high-stability image splicing display.

[0005] Therefore, there is an urgent need for a method that can still accurately perceive, correct, and adaptively converge to the splicing geometric error without repeated manual calibration, so as to improve the geometric consistency and long-term display stability of high-resolution TV wall image splicing. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, the purpose of this invention is to propose a method for assembling a high-resolution image splicing TV wall. This method aims to solve the technical problem of geometric misalignment at the splicing boundary, which is easily caused by existing splicing methods that rely on manual correction, especially when installing high-resolution, large-size TV walls with slight rotation angles.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for assembling a high-resolution image splicing television wall. The method for assembling a high-resolution image stitching video wall includes: Step S10: Acquire the entire curtain wall display image through the preset built-in image sensor, and perform the boundary structure feature extraction task based on the entire curtain wall display image using the self-excited structure perception driving mechanism of the display content, and output the boundary structure feature point set P; Step S20: Based on the boundary structure feature point set P, a geometric deviation inversion mechanism based on cross-screen boundary constraints is used to perform the cross-screen sub-pixel geometric deviation estimation task, and the geometric deviation parameter set is output. ; Step S30: Based on the geometric deviation parameter set A reverse mapping-driven mechanism is used to perform subpixel-level geometric correction of the display unit and output a geometrically corrected display unit image. Step S40: Based on the geometric correction display unit image, the geometric continuity consistency evaluation task is performed using the stitching boundary normal gradient consistency evaluation mechanism, and the stitching geometric consistency evaluation result is output; Step S50: For the geometrically corrected display unit image, an adaptive compensation mechanism based on residual gradient descent is used to perform the feedback correction task according to the splicing geometric consistency evaluation result, and the complete geometrically corrected display image is output.

[0008] Preferably, in step S10, the step of acquiring the entire curtain wall display image through a preset built-in image sensor, performing the boundary structure feature extraction task based on the entire curtain wall display image using a self-excited structure perception driving mechanism of the display content, and outputting the boundary structure feature point set P specifically includes: Step S101: Acquire the entire screen display image using a pre-set built-in image sensor. ,in, Indicates the pixel coordinate index along the horizontal direction of the curtain wall; This represents the pixel coordinate index along the vertical direction of the curtain wall; and determines the entire curtain wall display image based on pre-stored splicing topology information. Image of the i-th display unit With the image of the j-th display unit The public splicing boundary area; Step S102: Based on the common splicing boundary region, along the i-th display unit image With the image of the j-th display unit Extract pixel bands of a preset width along the normal direction of the splicing boundary, and output the boundary sub-image. ; Step S103: For the boundary sub-image The gradient operator is calculated using a local gray-level change calculation method based on central difference to obtain the horizontal gradient. gradient in the vertical direction ; and based on the horizontal gradient gradient in the vertical direction The gradient magnitude was calculated. , ; Step S104: Select gradient magnitude within the common splicing boundary region Greater than the preset gradient magnitude threshold The set of pixels is output as the set of boundary structure feature points P.

[0009] Preferably, in step S20, a cross-screen sub-pixel geometric deviation estimation task is performed based on the set of boundary structure feature points P using a geometric deviation inversion mechanism based on cross-screen boundary constraints, and the geometric deviation parameter set is output. The steps specifically include: Step S201: Extract the set of boundary structure feature points P in the image of the i-th display unit. With the image of the j-th display unit The first feature point sequence on both sides of the splicing boundary With the second feature point sequence ; Step S202: Based on the first feature point sequence With the second feature point sequence A two-dimensional rigid perturbation modeling method based on small-angle approximation is used to establish the geometric constraints of the two-dimensional rigid perturbation; the formula for the geometric constraints of the two-dimensional rigid perturbation is expressed as follows: ;in, Indicates the image in the j-th display unit. In the image, the horizontal pixel coordinates of the feature points after geometric perturbation mapping; To represent the image in the j-th display unit In the image, the vertical pixel coordinates of the feature points after geometric perturbation mapping; To represent the minute rotation angle between adjacent display units caused by the slight installation offset; To represent the sub-pixel level translation deviation along the horizontal direction of the splicing boundary; To represent the sub-pixel level translational deviation along the vertical direction of the splicing boundary; Step S203: Based on the first feature point sequence With the second feature point sequence The least squares method is used to construct the gray-level consistency error function of feature points. Under the premise of satisfying the two-dimensional rigid perturbation geometric constraints, the gray-level consistency error function of feature points is... Minimize the solution to obtain the set of geometric deviation parameters. .

[0010] Preferably, in step S30, based on the geometric deviation parameter set The steps of performing subpixel-level geometric correction of the display unit using a reverse mapping-driven mechanism and outputting the geometrically corrected display unit image specifically include: Step S301: Obtain the geometric deviation parameter set , To represent the minute rotation angle between adjacent display units caused by the slight installation offset; To represent the sub-pixel level translation deviation along the horizontal direction of the splicing boundary; To represent the sub-pixel level translational deviation along the vertical direction of the stitching boundary; based on the geometric deviation parameter set An inverse mapping function is constructed in the target display unit image using the principle of rigid body transformation inverse mapping. Inverse mapping function Used to describe the inverse mapping relationship between the corrected target pixel coordinates and the coordinates of consecutive source pixels in the original display unit image; Step S302: Obtain the target display unit image to be calibrated, and call the inverse mapping function constructed in step S301 for the target display unit image. The pixel coordinates in the image of the target display unit to be corrected Mapped to non-integer source pixel coordinates The mapping relationship is as follows: ;in, Indicates the x-coordinate of the non-integer source pixel. Represents the y-coordinate of a non-integer source pixel. This represents the horizontal coordinates of consecutive pixels in the target display unit image to be corrected; Represents the ordinate of consecutive horizontal pixels in the target display unit image to be corrected; non-integer source pixel coordinates. Used to characterize continuous geometric offsets caused by minute rotations and subpixel-level translations; Step S303: In the target display unit image to be corrected, for non-integer source pixel coordinates The pixel values ​​in the coordinate neighborhood are weighted using bicubic interpolation to obtain the corrected target pixel value, and finally the geometrically corrected display unit image is output.

[0011] Preferably, step S40, which involves performing a geometric continuity consistency evaluation task based on the geometrically corrected display unit image using a stitching boundary normal gradient consistency evaluation mechanism, and outputting the stitching geometric consistency evaluation result, specifically includes: Step S401: After generating the geometric correction display unit image, the entire curtain wall display image is redefined based on the pre-stored splicing topology information. Image of the i-th display unit With the image of the j-th display unit The common stitching boundary correction area is defined, and pixel bands of a preset width are selected on both sides of the stitching boundary along the boundary normal direction in the common stitching boundary correction area. A set of pixel pairs spanning the stitching boundary is then constructed according to the normal correspondence relationship. ; Step S402: Based on the set of pixel pairs Extract the first gray-level gradient component along the boundary normal direction in the common stitching boundary correction area. With the second gray-level gradient component ; and based on the first grayscale gradient component With the second gray-level gradient component Calculate the average gradient residual in the boundary normal direction ; Step S403: Based on the average gradient residual Generate and output the splicing geometric consistency evaluation results.

[0012] Preferably, in step S403, based on the average gradient residual The steps for generating and outputting the splicing geometric consistency evaluation results specifically include: calculating the average gradient residual... With the preset splicing geometric consistency threshold When comparing, When the current splicing boundary is determined to meet the geometric continuity and consistency requirements, the geometric consistency evaluation result is output as "passed"; when If the current splicing boundary is found to have residual geometric discontinuities, the geometric consistency evaluation result is "not passed".

[0013] Preferably, step S50, which involves performing a feedback correction task based on an adaptive compensation mechanism using residual gradient descent according to the splicing geometric consistency evaluation result for the geometrically corrected display unit image, and outputting a complete geometrically corrected display image, specifically includes: Step S501: When the splicing geometric consistency evaluation result is "passed", no correction processing is performed on the geometric correction display unit image; Step S502: When the splicing geometric consistency evaluation result is "not passed", the average gradient residual is used. As feedback, the image of the i-th display unit is constructed. With the image of the j-th display unit geometric parameter correction , ,in, This is a preset convergence coefficient used to control the magnitude of feedback correction. This indicates the feedback quantity relative to the geometric deviation parameter set. The gradient direction; Step S503: Adjust the geometric parameter amount The original geometric deviation parameter set is superimposed using a linear weighting method. Output optimized geometric deviation parameter set Based on the optimized geometric deviation parameter set Feedback is provided in steps S30 to S40 until the condition is met. The final output is a fully geometrically corrected display image.

[0014] The present invention also provides a high-resolution image splicing TV wall assembly system comprising: The image acquisition and boundary feature extraction module is used to acquire the entire curtain wall display image through a preset built-in image sensor, and to perform the boundary structure feature extraction task based on the entire curtain wall display image using a structure perception driving mechanism driven by the display content self-excitation, and output the boundary structure feature point set P; The geometric deviation inversion module is used to perform cross-screen sub-pixel geometric deviation estimation based on the set of boundary structure feature points P using a cross-screen boundary constraint-based geometric deviation inversion mechanism, and outputs a geometric deviation parameter set. ; Sub-pixel geometric correction module for use based on geometric deviation parameter set A reverse mapping-driven mechanism is used to perform subpixel-level geometric correction of the display unit and output a geometrically corrected display unit image. The splicing geometric consistency evaluation module is used to perform a geometric continuity consistency evaluation task based on the splicing boundary normal gradient consistency evaluation mechanism of the geometric correction display unit image, and output the splicing geometric consistency evaluation result; The adaptive feedback correction module is used to perform feedback correction tasks for the geometrically corrected display unit image by adopting an adaptive compensation mechanism based on residual gradient descent according to the splicing geometric consistency evaluation results, and outputting a complete geometrically corrected display image.

[0015] The present invention also provides a high-resolution image splicing TV wall assembly device, comprising: a memory, a processor, and a high-resolution image splicing TV wall assembly program stored in the memory and executable on the processor. When the high-resolution image splicing TV wall assembly program is executed by the processor, a high-resolution image splicing TV wall assembly method is implemented.

[0016] The present invention also provides a computer program product, including a high-resolution image splicing TV wall composition program, which, when executed by a processor, implements the high-resolution image splicing TV wall composition method.

[0017] The beneficial effects of this invention are as follows: By using a geometric deviation inversion mechanism based on the structural features of the splicing boundary, this invention can accurately estimate the minute rotational errors and sub-pixel translational errors caused by the micro-angle of installation between adjacent display units without the need for manual calibration and additional sensors. Combined with a sub-pixel geometric correction method driven by reverse mapping, it can perform fine correction on the display unit image, thereby effectively eliminating the boundary geometric misalignment problem caused by minute installation errors during the splicing of high-resolution TV walls, and significantly improving the continuity and consistency of the spliced ​​image at the geometric level.

[0018] Based on geometric correction, this invention introduces a geometric continuity evaluation mechanism based on the consistency of the normal gradient of the splicing boundary, and iteratively corrects the geometric deviation parameters through adaptive closed-loop feedback control of consistency residual gradient descent. This enables the splicing correction process to have quantifiable evaluation and convergent control capabilities, thereby avoiding the uncertainty and residual error accumulation problems of traditional one-time correction methods in complex splicing scenarios, and improving the display stability and splicing reliability of large-size high-resolution TV walls under long-term operation. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the first embodiment of a high-resolution image splicing television wall composition method according to the present invention.

[0021] Figure 2 This is a schematic diagram of the splicing display before feedback correction, representing a first embodiment of a high-resolution image splicing TV wall composition method of the present invention.

[0022] Figure 3 This is a schematic diagram of the splicing display after feedback correction, representing a first embodiment of a high-resolution image splicing TV wall composition method of the present invention.

[0023] Figure 4 This is a schematic diagram of the equipment used in the method for assembling a high-resolution image splicing TV wall according to the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1: As Figure 1 The diagram shown is a flowchart illustrating the first embodiment of the high-resolution image splicing TV wall composition method of the present invention, which presents the first embodiment of the high-resolution image splicing TV wall composition method of the present invention.

[0026] In the first embodiment, the method for assembling a high-resolution image splicing television wall includes: Step S10: Acquire the entire curtain wall display image through the preset built-in image sensor, and perform the boundary structure feature extraction task based on the entire curtain wall display image using the self-excited structure perception driving mechanism of the display content, and output the boundary structure feature point set P; It should be noted that the "self-excited structure perception driving mechanism of display content" refers to using the actual display content currently being played on the TV wall as the structural excitation source. By comprehensively analyzing the brightness changes, texture direction, contour transition and edge intersection relationships naturally formed in the splicing boundary area, boundary structural features that can stably reflect the spatial continuity relationship of adjacent display units are extracted.

[0027] It is understandable that, through the structure perception method of self-excitation of the display content described above, this step can continuously obtain structural information directly related to the geometric relationship of the splicing boundary under actual operation, so that the subsequent geometric deviation inversion process is based on the real display state, thereby effectively improving the stability and repeatability of cross-screen sub-pixel geometric deviation estimation, and avoiding significant interference to the splicing correction effect due to changes in display content or differences in environmental conditions.

[0028] Step S20: Based on the boundary structure feature point set P, a geometric deviation inversion mechanism based on cross-screen boundary constraints is used to perform the cross-screen sub-pixel geometric deviation estimation task, and the geometric deviation parameter set is output. ; It should be noted that the "geometric deviation inversion mechanism based on cross-screen boundary constraints" refers to, based on the obtained set of boundary structural feature points P, not treating each display unit as an independent image object, but rather using the splicing boundary between adjacent display units as the core of geometric constraints. By analyzing the deviations in spatial position, orientation distribution, and relative relationships of corresponding structural feature points located on both sides of the splicing boundary, the geometric deviation parameters introduced by installation errors between adjacent display units are derived in reverse. The geometric deviation parameter set includes at least rotation parameters for characterizing minute rotational relationships and sub-pixel-level translation parameters for characterizing the direction along the splicing boundary.

[0029] Understandably, by introducing a clear geometric constraint—the cross-screen stitching boundary—this step can confine the geometric deviation estimation process to a low-dimensional parameter space directly related to stitching continuity. This makes the inversion of sub-pixel-level geometric deviation no longer dependent on global image matching or complex high-degree-of-freedom models, thereby significantly improving the stability and convergence of the inversion process while ensuring estimation accuracy, and providing a reliable parameter basis for subsequent sub-pixel-level geometric correction.

[0030] It should be understood that, compared to traditional methods based on whole-image registration or relying on manual measurement of installation parameters, this step directly utilizes the structural feature information at the splicing boundary to perform geometric deviation inversion. This avoids uncertainties caused by changes in display content, interference from non-sponge areas, or manual calibration errors, making the geometric deviation estimation results more consistent with the actual splicing state. It is especially suitable for application scenarios in high-resolution TV walls where tiny installation deviations are difficult to perceive accurately by the naked eye or coarse-grained methods.

[0031] Step S30: Based on the geometric deviation parameter set A reverse mapping-driven mechanism is used to perform subpixel-level geometric correction of the display unit and output a geometrically corrected display unit image. It should be noted that the "reverse mapping driven mechanism" means that when performing geometric correction of the display unit image, the pixels in the original display image are not directly mapped to the corrected position. Instead, the corrected target pixel coordinates are used as the starting point, and the geometric deviation parameter set output in step S20 is combined to reverse determine the continuous source pixel coordinates in the original display unit image. The original image is then valued and reconstructed based on these continuous coordinates.

[0032] Understandably, when there are only minor rotation angles and subpixel-level translation errors between adjacent display units, the core objective of geometric correction is not to drastically transform the displayed image, but rather to finely adjust the pixel positions near the splicing boundary. By employing a subpixel-level geometric correction method driven by inverse mapping, this step can perform continuous and smooth geometric correction on the display unit image while maintaining the overall structure of the original display content. This allows the correction effect to transition naturally at the splicing boundary, thus providing a stable image foundation for subsequent evaluation of splicing geometric consistency.

[0033] Step S40: Based on the geometric correction display unit image, the geometric continuity consistency evaluation task is performed using the stitching boundary normal gradient consistency evaluation mechanism, and the stitching geometric consistency evaluation result is output; It should be noted that the "splicing boundary normal gradient consistency evaluation mechanism" refers to the following: after completing the geometric correction of the display unit image, the gray-level change characteristics on both sides of the splicing boundary are acquired along the normal direction of the splicing boundary region between adjacent display units. The degree of difference in the gray-level gradient components along the normal direction is then quantitatively analyzed to evaluate the geometric continuity of the splicing boundary. This evaluation mechanism focuses on the continuous transition of the geometric structure at the splicing boundary, rather than overall brightness or texture similarity.

[0034] Understandably, since the most direct visual manifestation of minute rotation angles or sub-pixel translation errors between display units is abrupt changes in structural direction or contour at the splicing boundary, performing a consistency analysis of the grayscale gradient along the normal direction of the splicing boundary can sensitively reflect whether the geometric correction is sufficient. The splicing geometric consistency evaluation results obtained through this step can serve as objective quantitative feedback on the effectiveness of the previous geometric correction step, providing a clear basis for whether further correction is needed.

[0035] It should be understood that, compared with traditional methods of evaluating splicing effects based on pixel grayscale difference, overall image similarity, or subjective manual observation, this step uses normal gradient consistency as an evaluation index, which effectively reduces the interference of changes in display content, uneven brightness, or texture differences on the evaluation results. This makes the evaluation results more focused on geometric continuity itself, significantly improving the accuracy and stability of splicing correction effect judgment from a technical perspective. It is especially suitable for application scenarios in high-resolution TV walls that are highly sensitive to minute geometric misalignments.

[0036] Step S50: For the geometrically corrected display unit image, an adaptive compensation mechanism based on residual gradient descent is used to perform the feedback correction task according to the splicing geometric consistency evaluation result, and the complete geometrically corrected display image is output.

[0037] It should be noted that the "adaptive compensation mechanism based on residual gradient descent" refers to using the splicing geometric consistency evaluation result output in step S40 as the feedback quantity, treating the geometric discontinuity residual at the splicing boundary as the optimization target, and gradually adjusting the geometric deviation parameters by analyzing the changing trend of this residual relative to the geometric deviation parameter set, so that the splicing geometric consistency evaluation index evolves in the direction of reduction. This mechanism dynamically corrects the compensation amount based on the current evaluation result in each feedback process, rather than using fixed or one-time set correction parameters.

[0038] Understandably, the installation errors of actual TV walls typically manifest as a superposition of minute rotation angles and sub-pixel level translation errors. These errors have cumulative and non-linear effects at the splicing boundaries, making it difficult to completely eliminate residual errors using only a single geometric correction. By introducing an adaptive compensation method based on residual gradient descent, this step can gradually approach the optimal state of splicing geometric continuity through multiple iterations, ensuring good convergence and stability of the geometric correction results. This guarantees that the final output geometrically corrected display image meets the expected consistency requirements at the splicing boundaries.

[0039] For example, such as Figure 2 and Figure 3 As shown, before correction, due to minute rotation angles and sub-pixel translation errors between adjacent display units, the structural lines at the splicing boundary exhibit significant misalignment and discontinuity, especially noticeable in oblique structures or high-contrast areas. After multiple rounds of feedback correction based on the splicing geometric consistency evaluation results, the structural contours on both sides of the splicing boundary gradually align, the geometric abrupt changes at the boundary are significantly reduced, and the misalignment that originally appeared at the splicing position is effectively eliminated. The overall displayed image presents a continuous and smooth visual effect in the splicing area. This comparative result intuitively demonstrates that this invention, by constructing an evaluation-driven closed-loop adaptive correction mechanism, can achieve sub-pixel-level geometric alignment of high-resolution TV wall spliced ​​images without manual intervention, thereby significantly improving the consistency and visual quality of the spliced ​​display.

[0040] Example 2: Furthermore, the high-resolution image splicing TV wall assembly system provided by the present invention employs a high-resolution image splicing TV wall assembly method from the above embodiments, which can solve the technical problem of assembling a high-resolution image splicing TV wall. Compared with the prior art, the beneficial effects of the high-resolution image splicing TV wall assembly system provided by the present invention are the same as the beneficial effects of the high-resolution image splicing TV wall assembly method provided by the above embodiments, and other technical features of the high-resolution image splicing TV wall assembly system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0041] Example 3: This invention provides a high-resolution image splicing TV wall assembly device, please refer to... Figure 4 A high-resolution image splicing TV wall assembly device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to perform a high-resolution image splicing TV wall assembly method as described in Embodiment 1 above. The high-resolution image splicing TV wall assembly device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. This high-resolution image splicing TV wall assembly device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this invention. The high-resolution image splicing TV wall assembly device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. Random access memory 1004 also stores various programs and data required for the operation of a high-resolution image splicing video wall assembly device. Processing device 1001, read-only memory 1002, and random access memory 1004 are interconnected via bus 1005. I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows a high-resolution image splicing video wall assembly device to exchange data wirelessly or via wired communication with other devices. Although a high-resolution image splicing video wall assembly device with various systems is shown in the figures, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0042] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for assembling a high-resolution image splicing video wall. The computer program product provided by this invention can solve the technical problem of assembling a high-resolution image splicing video wall. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as the beneficial effects of the high-resolution image splicing video wall assembly method provided in the above embodiments, and will not be repeated here.

[0043] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0044] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0045] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for assembling a high-resolution image splicing video wall, characterized in that, The methods include: Step S10: Acquire the entire curtain wall display image through the preset built-in image sensor, and perform the boundary structure feature extraction task based on the entire curtain wall display image using the self-excited structure perception driving mechanism of the display content, and output the boundary structure feature point set P; Step S20: Based on the boundary structure feature point set P, a geometric deviation inversion mechanism based on cross-screen boundary constraints is used to perform the cross-screen sub-pixel geometric deviation estimation task, and the geometric deviation parameter set is output. ; Step S30: Based on the geometric deviation parameter set A reverse mapping driving mechanism is used to perform subpixel-level geometric correction of the display unit and output a geometrically corrected display unit image. Step S40: Based on the geometric correction of the display unit image, the splicing boundary normal gradient consistency evaluation mechanism is used to perform the geometric continuity consistency evaluation task and output the splicing geometric consistency evaluation result; Step S50: For the geometrically corrected display unit image, an adaptive compensation mechanism based on residual gradient descent is used to perform the feedback correction task according to the splicing geometric consistency evaluation result, and the complete geometrically corrected display image is output.

2. The method for assembling a high-resolution image splicing video wall as described in claim 1, characterized in that, In step S10, the process of acquiring the entire curtain wall display image through a preset built-in image sensor, performing boundary structure feature extraction based on the entire curtain wall display image using a self-excited structure-aware driving mechanism for display content, and outputting the boundary structure feature point set P specifically includes: Step S101: Acquire the entire screen display image using a pre-set built-in image sensor. ,in, Indicates the pixel coordinate index along the horizontal direction of the curtain wall; This represents the pixel coordinate index along the vertical direction of the curtain wall; and determines the entire curtain wall display image based on pre-stored splicing topology information. Image of the i-th display unit With the image of the j-th display unit The public splicing boundary area; Step S102: Based on the common splicing boundary region, along the i-th display unit image With the image of the j-th display unit Extract pixel bands of a preset width along the normal direction of the stitching boundary, and output the boundary sub-image. ; Step S103: For the boundary sub-image The gradient operator is calculated using a local gray-level change calculation method based on central difference to obtain the horizontal gradient. gradient in the vertical direction ; and based on the horizontal gradient gradient in the vertical direction The gradient magnitude was calculated. , ; Step S104: Select gradient magnitude within the common splicing boundary region Greater than the preset gradient magnitude threshold The set of pixels is output as the set of boundary structure feature points P.

3. The method for assembling a high-resolution image splicing video wall as described in claim 2, characterized in that, In step S20, based on the boundary structure feature point set P, a geometric deviation inversion mechanism based on cross-screen boundary constraints is used to perform the cross-screen sub-pixel geometric deviation estimation task, and the geometric deviation parameter set is output. The steps specifically include: Step S201: Extract the set of boundary structure feature points P in the image of the i-th display unit. With the image of the j-th display unit The first feature point sequence on both sides of the splicing boundary With the second feature point sequence ; Step S202: Based on the first feature point sequence With the second feature point sequence A two-dimensional rigid perturbation modeling method based on small-angle approximation is used to establish the geometric constraints of the two-dimensional rigid perturbation; the formula for the geometric constraints of the two-dimensional rigid perturbation is expressed as follows: ;in, Indicates the image in the j-th display unit. In the image, the horizontal pixel coordinates of the feature points after geometric perturbation mapping; To represent the image in the j-th display unit In the image, the vertical pixel coordinates of the feature points after geometric perturbation mapping; To represent the minute rotation angle between adjacent display units caused by the slight installation offset; To represent the sub-pixel level translation deviation along the horizontal direction of the splicing boundary; To represent the sub-pixel level translational deviation along the vertical direction of the splicing boundary; Step S203: Based on the first feature point sequence With the second feature point sequence The least squares method is used to construct the gray-level consistency error function of feature points. Under the premise of satisfying the two-dimensional rigid perturbation geometric constraints, the gray-level consistency error function of feature points is... Minimize the solution to obtain the set of geometric deviation parameters. .

4. The method for assembling a high-resolution image splicing video wall as described in claim 1, characterized in that, In step S30, based on the geometric deviation parameter set The steps of performing subpixel-level geometric correction of the display unit using a reverse mapping-driven mechanism and outputting the geometrically corrected display unit image specifically include: Step S301: Obtain the geometric deviation parameter set , To represent the minute rotation angle between adjacent display units caused by the slight installation offset; To represent the sub-pixel level translation deviation along the horizontal direction of the splicing boundary; To represent the sub-pixel level translational deviation along the vertical direction of the stitching boundary; based on the geometric deviation parameter set An inverse mapping function is constructed in the target display unit image using the principle of rigid body transformation inverse mapping. Inverse mapping function Used to describe the inverse mapping relationship between the corrected target pixel coordinates and the coordinates of consecutive source pixels in the original display unit image; Step S302: Obtain the target display unit image to be calibrated, and call the inverse mapping function constructed in step S301 for the target display unit image. The pixel coordinates in the image of the target display unit to be corrected Mapped to non-integer source pixel coordinates The mapping relationship is as follows: ;in, Indicates the x-coordinate of the non-integer source pixel. Represents the y-coordinate of a non-integer source pixel. This represents the horizontal coordinates of consecutive pixels in the target display unit image to be corrected; Represents the ordinate of consecutive horizontal pixels in the target display unit image to be corrected; non-integer source pixel coordinates. Used to characterize continuous geometric offsets caused by minute rotations and subpixel-level translations; Step S303: In the target display unit image to be corrected, for non-integer source pixel coordinates The pixel values ​​in the coordinate neighborhood are weighted using bicubic interpolation to obtain the corrected target pixel value, and finally the geometrically corrected display unit image is output.

5. The method for assembling a high-resolution image splicing video wall as described in claim 2, characterized in that, Step S40, which involves performing a geometric continuity consistency evaluation task based on the geometrically corrected display unit image using a stitching boundary normal gradient consistency evaluation mechanism, and outputting the stitching geometric consistency evaluation result, specifically includes: Step S401: After generating the geometric correction display unit image, the entire curtain wall display image is redefined based on the pre-stored splicing topology information. Image of the i-th display unit With the image of the j-th display unit The common stitching boundary correction area is defined, and pixel bands of a preset width are selected on both sides of the stitching boundary along the boundary normal direction in the common stitching boundary correction area. A set of pixel pairs spanning the stitching boundary is then constructed according to the normal correspondence relationship. ; Step S402: Based on the set of pixel pairs Extract the first gray-level gradient component along the boundary normal direction in the common stitching boundary correction area. With the second gray-level gradient component ; and based on the first grayscale gradient component With the second gray-level gradient component Calculate the average gradient residual in the boundary normal direction ; Step S403: Based on the average gradient residual Generate and output the splicing geometric consistency evaluation results.

6. The method for assembling a high-resolution image splicing video wall as described in claim 5, characterized in that, In step S403, based on the average gradient residual The steps for generating and outputting the splicing geometric consistency evaluation results specifically include: calculating the average gradient residual... With the preset splicing geometric consistency threshold When comparing, When the current splicing boundary is determined to meet the geometric continuity and consistency requirements, the geometric consistency evaluation result is output as "passed"; when If the current splicing boundary is found to have residual geometric discontinuity, the geometric consistency evaluation result is "not passed".

7. The method for assembling a high-resolution image splicing video wall as described in claim 5, characterized in that, Step S50, for the geometrically corrected display unit image, involves performing a feedback correction task using an adaptive compensation mechanism based on residual gradient descent, according to the splicing geometric consistency evaluation result, and outputting a complete geometrically corrected display image. This step specifically includes: Step S501: When the splicing geometric consistency evaluation result is "passed", no correction processing is performed on the geometric correction display unit image; Step S502: When the splicing geometric consistency evaluation result is "not passed", the average gradient residual is used. As feedback, the image of the i-th display unit is constructed. With the image of the j-th display unit Geometric parameter correction , ,in, This is a preset convergence coefficient used to control the magnitude of feedback correction. This indicates the feedback quantity relative to the geometric deviation parameter set. The gradient direction; Step S503: Adjust the geometric parameter amount The original geometric deviation parameter set is superimposed using a linear weighting method. Output optimized geometric deviation parameter set Based on the optimized geometric deviation parameter set Feedback is provided in steps S30 to S40 until the condition is met. The final output is a fully geometrically corrected display image.

8. A high-resolution image splicing video wall assembly system, applied to the high-resolution image splicing video wall assembly method according to any one of claims 1 to 7, characterized in that, The high-resolution image splicing video wall system comprises: The image acquisition and boundary feature extraction module is used to acquire the entire curtain wall display image through a preset built-in image sensor, and to perform the boundary structure feature extraction task based on the entire curtain wall display image using a structure perception driving mechanism driven by the display content self-excitation, and output the boundary structure feature point set P; The geometric deviation inversion module is used to perform cross-screen sub-pixel geometric deviation estimation based on the set of boundary structure feature points P using a cross-screen boundary constraint-based geometric deviation inversion mechanism, and outputs a geometric deviation parameter set. ; Sub-pixel geometric correction module for use based on geometric deviation parameter set A reverse mapping driving mechanism is used to perform subpixel-level geometric correction of the display unit and output a geometrically corrected display unit image. The splicing geometric consistency evaluation module is used to perform a geometric continuity consistency evaluation task based on the splicing boundary normal gradient consistency evaluation mechanism of the geometric correction display unit image, and output the splicing geometric consistency evaluation result; The adaptive feedback correction module is used to perform feedback correction tasks for the geometrically corrected display unit image by adopting an adaptive compensation mechanism based on residual gradient descent according to the splicing geometric consistency evaluation results, and outputting a complete geometrically corrected display image.

9. A high-resolution image splicing video wall assembly device, characterized in that, The high-resolution image splicing TV wall assembly device includes: a memory, a processor, and a high-resolution image splicing TV wall assembly program stored in the memory and executable on the processor. When the high-resolution image splicing TV wall assembly program is executed by the processor, it implements a high-resolution image splicing TV wall assembly method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a high-resolution image splicing TV wall assembly program, which, when executed by a processor, implements a high-resolution image splicing TV wall assembly method according to any one of claims 1 to 7.

Images