Projector control method and projector

Abstract

The invention relates to the technical field of projectors, in particular to a projector control method and a projector. The method comprises the following steps: performing image projection signal processing on target projection image data to generate initial projection picture data; performing projection point cloud space-time registration on the initial projection picture data by using the dual laser radar assembly to obtain projection scanning point cloud space data; performing shielding area geometric feature analysis on the projection scanning point cloud spatial data, determining an ideal projection area, and generating ideal rectangular projection area data; performing intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters; and performing projection attitude control instruction processing on the intelligent projection control parameters to obtain projection control optimization effect data. According to the invention, the projection control parameters are automatically adjusted through intelligent identification of the projection surface, so that the projector can maintain a stable projection effect in different environments, and the quality of a projection image is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of projectors, and in particular to a projector control method and a projector. Background Art

[0002] Projection display technology is widely used in various scenarios, such as business presentations, education and teaching, and home entertainment. Users have higher and higher requirements for the quality and convenience of projection display. The uncertainty of the projection environment has become a challenge. These environments include projection screens of different positions, shapes, and sizes, as well as obstacles and light interference. For example, the use environment of the projector is different in home, education, conference, and exhibition, which makes the setting of the projector complicated. For these complex scenarios, limited by the differences in the position and shape of the projection screen and the interference of obstacles in the projection path, for example, in some conference rooms or exhibition halls, there may be obstacles such as pillars, tables and chairs in the projection area, which will cause the projection image to be missing or deformed, affecting the viewing experience. However, the traditional projector projection method mainly relies on manual adjustment of the projection angle and focal length to achieve trapezoidal correction. This method is cumbersome, inefficient, and difficult to accurately calibrate. Especially when the projection screen is irregular in shape or there are obstructions, manual adjustment is more difficult, which can easily lead to missing, deformed or incorrect correction of the image, seriously affecting the projection effect. Summary of the invention

[0003] Based on this, the present invention provides a projector control method and a projector to solve at least one of the above technical problems.

[0004] To achieve the above object, a projector control method comprises the following steps:

[0005] Step S1: acquiring target projection image data; the image processor processes the target projection image data for image projection signals, and controls the projection lens assembly to perform image projection, generating initial projection screen data; using the dual laser radar assembly to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining projection scanning point cloud spatial data;

[0006] Step S2: extracting projection surface boundary points from the projection scanning point cloud spatial data to generate projection plane boundary point cloud data; processing projection surface geometric features according to the projection plane boundary point cloud data to generate projection surface geometric feature data;

[0007] Step S3: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data; performing occlusion area geometric feature analysis on the projection picture contour data through the projection surface geometric feature data to generate occlusion area geometric feature data; determining an ideal projection area according to the occlusion area geometric feature data to generate ideal rectangular projection area data; performing intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters;

[0008] Step S4: Process the projection posture control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; evaluate the projection control effect based on the real-time projection image data to obtain projection control optimization effect data.

[0009] The present invention effectively obtains the spatial data in the projection area through the generation of the initial projection picture and the precise point cloud registration of the dual laser radar, ensuring that the projection effect can be intelligently adjusted under the conditions of different positions, sizes and shapes of the projection screen. Through further extraction of projection surface boundary points and geometric feature processing, the contour features of the complex projection surface can be identified and adapted, so that the projection picture is more closely aligned with the actual projection surface. Even when the projection screen is irregular or there are obstacles, it can be automatically corrected to achieve the integrity of the projection area and distortion-free projection of the image. Then, by intelligently analyzing the geometric feature data of the occluded area, the optimal projection area can be flexibly determined, thereby avoiding the phenomenon of missing images caused by obstacles and ensuring the consistency of the viewing experience. The generated projection control parameters are combined with the collection and feedback evaluation of real-time images, so that the projector can perform intelligent projection posture adjustment and dynamically optimize the display effect according to environmental changes, so that the projector can maintain a stable correction effect in different environments without manual adjustment, significantly improving the projection accuracy and the automatic adaptability of the projection effect. The intelligent and real-time response characteristics of this solution are particularly suitable for complex projection needs in various scenarios such as home, education, and business presentations. It successfully solves the problems of cumbersome operation and inaccurate correction of the traditional manual adjustment method, making the projection operation more convenient and efficient, and significantly improving the application flexibility and user experience of the projector. Therefore, a projector control method of the present invention uses a dual laser radar component to perform high-precision spatial scanning of the projection area, analyze the geometric features of the projection surface, and automatically optimize the posture and image correction parameters of the projector to adapt to various complex and changeable projection environments, thereby achieving real-time adjustment of projection parameters and ensuring clear and distortion-free projection effects on projection surfaces of different positions, shapes, and sizes.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: acquiring target projection image data; the image processor processes the target projection image data into image projection signals, and controls the projection lens assembly to perform image projection, thereby generating initial projection picture data;

[0012] Step S12: Based on the preset scanning frequency data, the dual laser radar components are used to perform a left and right dual-path horizontal scanning of the projection area on the initial projection image data to generate left and right projection horizontal scanning data;

[0013] Step S13: using a servo motor to drive the dual laser radar assembly to perform vertical interval layered scanning of the projection area to generate vertical layered scanning data of the projection area;

[0014] Step S14: performing point cloud noise filtering on the left and right projection horizontal scanning data and the projection area vertical layered scanning data to obtain projection horizontal scanning point cloud data and projection vertical scanning point cloud data respectively;

[0015] Step S15: performing point cloud density equalization processing according to the projected horizontal scanning point cloud data and the projected vertical scanning point cloud data, and performing spatiotemporal registration of the projected point cloud to obtain projected scanning point cloud spatial data.

[0016] The present invention ensures that the projection system has a highly consistent projection basis in different environments through precise image projection signal processing and the generation of the initial projection picture. The left and right horizontal scanning of the dual laser radar based on the preset scanning frequency effectively divides the projection area into left and right scanning channels, so that the details of the horizontal area of ​​the projection surface can be fully captured. The radar component is then controlled by a servo motor to realize vertical interval layered scanning, which further refines the point cloud distribution in the vertical direction and ensures the full capture of the depth information of the projection surface. This combination of horizontal and vertical bidirectional scanning makes the spatial feature recognition of the projection area more accurate and avoids the loss of information caused by complex environments or obstacles. By filtering the noise points of the horizontal and vertical scanning data, unnecessary environmental interference is eliminated and the purity of the point cloud data is improved. Finally, the point cloud density equalization processing ensures the uniform spatial distribution of the scanned point cloud, thereby achieving high-precision spatiotemporal registration of the projection picture.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: performing rough plane fitting of the projection scanning point cloud spatial data to generate initial fitting projection plane parameters;

[0019] Step S22: performing iterative weighted parameter optimization on the initial fitting projection plane parameters to obtain optimized fitting projection plane parameters;

[0020] Step S23: using the projection scanning point cloud spatial data to perform point cloud mapping processing on the optimized fitting projection plane parameters, and extracting the projection surface boundary points to generate projection plane boundary point cloud data;

[0021] Step S24: performing boundary point clustering processing on the projection plane boundary point cloud data, and performing boundary point cluster polygon fitting to obtain projection plane polygon boundary data;

[0022] Step S25: performing projection plane corner point recognition on the projection plane boundary point cloud data to obtain projection plane corner point coordinate data;

[0023] Step S26: Perform projection surface geometric feature processing on the projection plane corner point coordinate data through the projection plane polygon boundary data to generate projection surface geometric feature data.

[0024] The present invention effectively obtains the initial geometric features of the projection area by rough plane fitting of the projection scanning point cloud spatial data. By iterative weighted optimization parameters, the accuracy of the projection plane fitting is further improved, so that the projection plane parameters can fit the actual environment more accurately, especially when facing irregular or inclined projection surfaces, ensuring the flatness and integrity of the projection picture. Subsequently, point cloud mapping processing and boundary point extraction accurately identify the boundary contour of the projection surface, greatly improving the adaptive ability to complex boundary shapes. Boundary point clustering processing and polygon fitting further convert the boundary data into continuous and clear polygon boundaries, which can achieve accurate picture display on irregular projection screens and avoid boundary distortion. Through corner point recognition, not only can the corner point coordinates of the projection plane be captured, but also accurate positioning on the polygon boundary is achieved. This accurate boundary recognition ensures complete coverage of the projection picture and clear presentation of the edge. By processing the geometric features of polygon boundary data and corner point coordinate data, complete projection surface geometric feature data is generated, enhancing the adaptability of the projection system to complex shape projection surfaces.

[0025] Preferably, step S3 comprises the following steps:

[0026] Step S31: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data;

[0027] Step S32: performing geometric feature analysis of the occlusion area on the projection picture contour data using the projection surface geometric feature data to generate geometric feature data of the occlusion area;

[0028] Step S33: performing projection occlusion spatial distribution analysis on the occlusion area geometric feature data using the projection surface geometric feature data to obtain projection occlusion spatial distribution data;

[0029] Step S34: Determine the ideal projection area according to the projection occlusion spatial distribution data to generate ideal rectangular projection area data; perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters.

[0030] The present invention obtains the contour data of the initial projection picture quickly and accurately through the projection picture contour recognition. Combined with the projection surface geometric feature data, the geometric feature analysis of the occluded area in the picture can effectively identify and distinguish the occluded area in the picture. Based on the spatial distribution analysis of the occluded area, the specific position and range of each occluded area are accurately located, and detailed occluded spatial distribution data are provided. These data provide a basis for determining the ideal projection area, so that the control system can avoid the occluded area and intelligently select a rectangular projection area suitable for display, thereby ensuring the integrity of the picture and the projection quality. Through the generation of intelligent projection control parameters, the dynamic optimization control of the projection picture is further realized, so that the projection area can be flexibly adjusted and automatically adapted to the complex spatial distribution in different environments. It not only realizes the intelligent avoidance of the occluded area and the selection of the adaptive projection area, but also improves the visibility and overall look and feel of the projection picture, and solves the problem of picture missing and deformation caused by obstacles.

[0031] Preferably, step S32 includes the following steps:

[0032] Step S321: performing actual projection contour difference analysis on the projection picture contour data through the projection surface geometric feature data to generate actual projection difference contour data;

[0033] Step S322: performing potential occluder point cloud recognition on the projection scanning point cloud space data based on the actual projection difference contour data to obtain potential occluder point cloud data;

[0034] Step S323: clustering the potential occluder point cloud data into occluded area point cloud data to generate occluder clustered spatial area data;

[0035] Step S324: extracting the boundary of the occluder according to the occluder clustering spatial region data to generate occluder region boundary data;

[0036] Step S325: classifying the occlusion region boundary data according to the shape of the occlusion object to obtain occlusion region shape classification data;

[0037] Step S326: performing geometric feature analysis of the occlusion region according to the occlusion region shape classification data and the occlusion region boundary data to generate geometric feature data of the occlusion region.

[0038] The present invention accurately identifies the deviation between the actual projection picture and the expected projection profile by performing differential analysis on the projection picture profile data and the projection surface geometric feature data, and generates the actual projection differential profile data. Based on the differential profile data, the projection scanning point cloud data is further identified for potential obstructions, so as to accurately find out the obstacles that affect the projection effect. The effective execution of this step can actively detect and avoid any interference that affects the projection area, and avoid the missing or deformation of the projection picture caused by obstacles. By clustering the potential obstruction point cloud data, the specific distribution area of ​​these obstacles in the projection space is identified, so that the influence range of each obstruction is clearer, ensuring that the obstruction area can be accurately identified and distinguished from other areas, avoiding the occurrence of misjudgment and missed judgment. By classifying the shape of the obstruction area boundary, the nature and position of different types of obstructions can be understood more intelligently, and then the most effective avoidance strategy can be formulated. The in-depth analysis of the geometric features of the obstruction area provides comprehensive geometric data support for intelligent projection control, ensuring the maximization of the projection effect and the integrity of the image.

[0039] Preferably, step S34 includes the following steps:

[0040] Step S341: Calculate the projection boundary adjustment distance according to the projection occlusion space distribution data to obtain projection boundary adjustment distance data;

[0041] Step S342: performing adjustment boundary mapping on the projection occlusion space distribution data through the projection boundary adjustment distance data to generate target projection adjustment boundary data;

[0042] Step S343: using the target projection adjustment boundary data to identify the non-occluded overlapping area of ​​the occluded area geometric feature data, and performing a projection area visibility evaluation to obtain area visibility evaluation data;

[0043] Step S344: Based on the target projection adjustment boundary data, the projection boundary adjustment distance data is corrected by using the regional visibility evaluation data to generate optimal projection adjustment boundary data;

[0044] Step S345: estimating the projector position offset of the projection surface geometric feature data through the optimal projection adjustment boundary data, and performing gradient descent optimization to generate projector position adjustment data;

[0045] Step S346: determining the ideal projection area according to the projector position adjustment data to generate ideal rectangular projection area data; performing projection image correction control processing on the ideal rectangular projection area data to generate projection image correction control data;

[0046] Step S347: combining the projector position adjustment data and the projection image correction control data into projector intelligent control parameters to obtain intelligent projection control parameters.

[0047] The present invention calculates the projection boundary adjustment distance through the projection occlusion spatial distribution data. This adjustment distance data enables the controller to clearly understand the actual boundary of the projection area, and generates the target projection adjustment boundary data by adjusting the boundary mapping, so as to ensure that the projection area meets the spatial distribution requirements of the actual environment. By analyzing the geometric feature data of the occlusion area, the overlapping area without occlusion can be identified, and the regional visibility evaluation is performed to accurately evaluate the projection effect and visibility of each area. Based on these evaluation results, the projection boundary adjustment distance is corrected, and the optimal projection adjustment boundary data is generated to ensure the maximum coverage and optimal display effect of the projection area. By applying the optimal projection adjustment boundary data, by estimating the projector position offset and performing gradient descent optimization, the accurate adjustment of the projector position is achieved, the optimization of the projection angle and focal length of the projector is ensured, and unnecessary picture distortion or misalignment problems are avoided. Based on the new projector position adjustment data, the ideal projection area is re-determined, and the correction control of the projected image is performed to ensure that the picture is still clear and distortion-free in different environments. Finally, these adjustment data are combined into intelligent projection control parameters to ensure that the projector can automatically adapt to any complex environment and achieve accurate projection effects.

[0048] Preferably, step S346 includes the following steps:

[0049] Step S3461: determining the ideal projection area according to the projector position adjustment data, and generating ideal rectangular projection area data;

[0050] Step S3462: performing projection area triangulation on the ideal rectangular projection area data to generate ideal projection sub-area division data;

[0051] Step S3463: constructing a projection sub-region topological relationship for the ideal projection sub-region division data to obtain projection sub-region topological data;

[0052] Step S3464: mapping the projection sub-region corner points according to the projection sub-region topology data, and calculating the local perspective transformation matrix to generate local perspective transformation matrix data;

[0053] Step S3465: Perform light path projection processing on the local perspective transformation matrix data to generate a light path projection angle matrix; perform projection image correction control processing according to the light path projection angle matrix to generate projection image correction control data.

[0054] The present invention determines an ideal rectangular projection area according to the projector position adjustment data, thereby ensuring that the projection area meets actual needs and avoiding image distortion or dislocation caused by projector position deviation. The ideal rectangular projection area is triangulated and the area is divided into multiple sub-areas, effectively subdividing the large-scale projection area into smaller and easier-to-handle parts. This operation makes the projection adjustment more precise and can adapt to projection surfaces of different shapes and sizes. The topological relationship between the projection sub-areas is established according to the sub-area division, and the spatial and positional relationship between the sub-areas is clarified. Through further processing of the sub-area topological data, corner point mapping can be performed and a local perspective transformation matrix can be generated, which effectively handles the geometric deformation of the projection surface caused by the angle change and ensures the accurate presentation of the projection image in different sub-areas. Finally, through the calculation of the light path projection angle matrix, the change of the light path during the projection process is successfully simulated, and based on this, the projection image is corrected to correct the image distortion caused by the projection angle or shape change.

[0055] Preferably, step S3465 includes the following steps:

[0056] Predicting projection distortion according to the projector position adjustment data to obtain projection distortion prediction data;

[0057] Perform light path projection processing according to the local perspective transformation matrix data to generate a light path projection angle matrix;

[0058] Based on the projection distortion prediction data, the sub-region distortion compensation amount is calculated for the projection sub-region topology data using the light path projection angle matrix to generate the sub-region deformation compensation amount data;

[0059] Remapping the projection pixels according to the sub-region deformation compensation amount data, and performing projection image scaling processing to generate image projection reconstruction data;

[0060] Performing projection area mask processing on the ideal rectangular projection area data through image projection reconstruction data to generate projection area mask data;

[0061] A projection image correction control process is performed based on the projection area mask data and the image projection reconstruction data to generate projection image correction control data.

[0062] The present invention predicts projection distortion through projector position adjustment data, can accurately predict the image distortion that occurs, avoids the visual inaccuracy problem caused by projection angle or space limitation during projection, uses local perspective transformation matrix data to perform light path projection processing, generates a light path projection angle matrix, and accurately simulates the changes of light paths in different areas. Based on the projection distortion prediction data and the light path projection angle matrix, the topological data of the projection sub-area is calculated for distortion compensation, effectively corrects the geometric distortion of each sub-area during the projection process, and ensures that the image of each sub-area can remain accurate and without deformation. The image is pixel remapped and scaled according to the compensation amount data, so that the image in the projection area can be presented without distortion. By performing projection area mask processing on the image projection reconstruction data, the effective projection area and the invalid area are effectively isolated, ensuring that the image is presented only in the actual visible area. The projection image correction control processing is performed using the projection area mask data and the image projection reconstruction data, ensuring that the projector can automatically adjust and eliminate any image distortion caused by the projection angle or occlusion, thereby maintaining the clarity and accuracy of the picture in a complex environment.

[0063] Preferably, step S4 comprises the following steps:

[0064] Step S41: Processing the intelligent projection control parameters into projection posture control instruction to obtain optimized projection control instruction data;

[0065] Step S42: executing the motor control instruction according to the optimized projection control instruction data, and using the image sensor to collect the real-time projection picture to obtain the real-time projection picture data;

[0066] Step S43: performing projection reference mark recognition on the real-time projection screen data through the preset projection reference mark data, and extracting key feature points of the projection screen to generate optimized projection screen feature point data;

[0067] Step S44: Calculate the geometric distortion of the optimized projection picture feature point data to generate optimized geometric distortion data; evaluate the edge clarity of the real-time projection picture data to obtain edge clarity data; evaluate the brightness uniformity of the real-time projection picture data to obtain brightness uniformity data; evaluate the image projection integrity of the real-time projection picture data to obtain image projection integrity data;

[0068] Step S45: performing color restoration evaluation on the real-time projection picture data and the initial projection picture data based on the optimized projection picture feature point data to obtain color restoration data;

[0069] Step S46: using the preset projection control optimization weight data to perform a weighted evaluation of the control effect on the optimized geometric distortion data, edge clarity data, brightness uniformity data, image projection integrity data and color restoration data, and obtaining projection control optimization effect data.

[0070] The present invention optimizes the projection posture control through intelligent projection control parameters, so that the projection system can adapt to changes in different angles and positions in real time, ensuring that the projection effect is always stable. By performing motor execution control on the optimized control instructions and combining the picture data collected in real time by the image sensor, the latest projection status information can be continuously monitored and obtained during the projection process. In the real-time picture data collection, based on the preset projection reference mark recognition and key feature point extraction, the key features of the projection picture are accurately obtained, making the geometric distortion control of the projection image more detailed. Multi-dimensional evaluation is performed on indicators such as edge clarity, brightness uniformity and picture integrity to ensure that the overall quality of the picture meets the ideal visual effect. By evaluating the color reproduction of the real-time picture and the data of the initial projection image, the color expression of the projection image is further guaranteed, so that the picture can maintain color consistency and naturalness under different ambient light. Through the preset optimization weights, multiple evaluation data such as geometric distortion, edge clarity, brightness uniformity, projection integrity and color reproduction are weighted and comprehensively processed, effectively evaluating the overall performance of the projection effect, and significantly improving the intelligence and automation level of the projection system.

[0071] Preferably, the present invention further provides a projector, the projector comprising a projection host and a dual laser radar assembly, the projection host is provided with a projection lens assembly, an image processor and a main controller, the dual laser radar assembly comprises a first laser radar, a second laser radar and two servo motors, the first laser radar and the second laser radar are respectively fixed on both sides of the projection host, and are driven by the servo motors to perform vertical scanning, the dual laser radar assembly, the projection lens assembly and the image processor are electrically connected to the main controller through a data bus, wherein the main controller is used to execute the projector control method as described above, and the main controller comprises the following modules:

[0072] The projection point cloud processing module is used to obtain the target projection image data; the image processor processes the target projection image data with image projection signals, and controls the projection lens assembly to project the image, generating initial projection screen data; the dual laser radar assembly is used to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining the projection scanning point cloud spatial data;

[0073] The projection surface feature extraction module is used to extract the projection surface boundary points from the projection scanning point cloud space data to generate the projection plane boundary point cloud data; perform projection surface geometric feature processing based on the projection plane boundary point cloud data to generate the projection surface geometric feature data;

[0074] The projection control parameter optimization module is used to identify the projection picture contour of the initial projection picture data to obtain the projection picture contour data; to analyze the geometric characteristics of the occlusion area of ​​the projection picture contour data through the projection surface geometric characteristic data to generate the geometric characteristic data of the occlusion area; to determine the ideal projection area according to the geometric characteristic data of the occlusion area to generate the ideal rectangular projection area data; to perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain the intelligent projection control parameters;

[0075] The projection attitude control module is used to process the projection attitude control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; the projection control effect is evaluated based on the real-time projection image data to obtain projection control optimization effect data. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 A schematic diagram of the steps of a projector control method of the present invention;

[0077] Figure 2 for Figure 1 Detailed implementation steps of step S1 in FIG.

[0078] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0079] Figure 4 Frame the square frame of the projector;

[0080] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0081] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0082] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0083] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0084] To achieve this, please refer to Figures 1 to 4 The present invention provides a projector control method, comprising the following steps:

[0085] Step S1: acquiring target projection image data; the image processor processes the target projection image data for image projection signals, and controls the projection lens assembly to perform image projection, generating initial projection screen data; using the dual laser radar assembly to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining projection scanning point cloud spatial data;

[0086] Step S2: extracting projection surface boundary points from the projection scanning point cloud spatial data to generate projection plane boundary point cloud data; processing projection surface geometric features according to the projection plane boundary point cloud data to generate projection surface geometric feature data;

[0087] Step S3: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data; performing occlusion area geometric feature analysis on the projection picture contour data through the projection surface geometric feature data to generate occlusion area geometric feature data; determining an ideal projection area according to the occlusion area geometric feature data to generate ideal rectangular projection area data; performing intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters;

[0088] Step S4: Process the projection posture control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; evaluate the projection control effect based on the real-time projection image data to obtain projection control optimization effect data.

[0089] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of the steps of the projector control method of the present invention. In this embodiment, the projector control method includes the following steps:

[0090] Step S1: acquiring target projection image data; the image processor processes the target projection image data for image projection signals, and controls the projection lens assembly to perform image projection, generating initial projection screen data; using the dual laser radar assembly to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining projection scanning point cloud spatial data;

[0091] In an embodiment of the present invention, target projection image data is obtained, for example, a picture or video file to be projected is read from a computer, a mobile device or other data source. The image processor receives these data and performs image projection signal processing, such as adjusting the image resolution, color space, brightness and contrast, etc., to adapt to the characteristics of the projector and the projection environment. The processed image data is sent to the projection lens assembly, which controls it to perform image projection and generate an initial projection screen. At the same time, the dual laser radar assembly starts working, scans the projection area at a preset scanning frequency, emits a laser beam and receives a reflected signal. By analyzing the flight time and intensity of the laser beam, the three-dimensional point cloud data of the projection area can be obtained. These point cloud data represent the position information of the projection screen in three-dimensional space. In order to make the point cloud data correspond to the projection screen image data, point cloud spatiotemporal registration is required. This can be achieved by embedding a known pattern or mark in the projection screen, or by using image processing technology to identify the feature points of the projection screen, thereby establishing a spatial correspondence between the point cloud data and the image data, and finally obtaining the projection scanning point cloud space data.

[0092] Step S2: extracting projection surface boundary points from the projection scanning point cloud spatial data to generate projection plane boundary point cloud data; processing projection surface geometric features according to the projection plane boundary point cloud data to generate projection surface geometric feature data;

[0093] In an embodiment of the present invention, the projection scanning point cloud spatial data obtained in step S1 is used to perform rough plane fitting of the point cloud using a RANSAC (random sampling consensus) algorithm to obtain initial fitting projection plane parameters, such as the plane normal vector and the distance between the plane and the origin. In order to improve the fitting accuracy, the iterative weighted least squares method is used to iteratively optimize the initial fitting projection plane parameters so that the fitting plane can better adapt to the distribution of the projection point cloud data. The projection scanning point cloud spatial data is subjected to point cloud mapping processing using the optimized projection plane parameters, the point cloud data is projected onto the fitting plane, and the projection plane boundary points are extracted by an edge detection algorithm based on the distribution of the point cloud on the plane to generate projection plane boundary point cloud data. Then, DBSCAN (density-based spatial clustering for noise applications) clustering processing is performed on the boundary point cloud data to divide the boundary points into several clusters. Polygon fitting is performed on each point cluster, for example, the least squares method is used to fit the polygon to obtain projection plane polygon boundary data. The projection plane corner points are identified by a corner detection algorithm, such as the Harris corner detection algorithm, to obtain projection plane corner point coordinate data. According to the projection plane polygon boundary data and corner point coordinate data, the geometric features of the projection surface, such as area, perimeter, aspect ratio, angle, etc., are calculated to generate the projection surface geometric feature data.

[0094] Step S3: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data; performing occlusion area geometric feature analysis on the projection picture contour data through the projection surface geometric feature data to generate occlusion area geometric feature data; determining an ideal projection area according to the occlusion area geometric feature data to generate ideal rectangular projection area data; performing intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters;

[0095] In an embodiment of the present invention, edge detection and contour extraction are performed on the initial projection picture data generated in step S1, for example, edge detection is performed using the Canny operator, and then the projection picture contour is extracted using a morphological processing method to obtain projection picture contour data. The projection surface geometric feature data generated in step S2 is combined with the projection picture contour data to perform occlusion area analysis on the projection picture. By comparing the projection picture contour with the projection surface boundary, the occluded area in the projection picture is identified, and the spatial position and shape of the occluder are determined using the point cloud data. Point cloud clustering is performed on the identified occlusion area, the boundary of the occlusion area is extracted, and shape classification (such as rectangle, circle, irregular shape, etc.) is performed, and finally the occlusion area geometric feature data is generated. According to the occlusion area geometric feature data and the projection surface geometric feature data, the ideal projection area is determined. The ideal projection area is usually a rectangular area, and its size and position can maximize the use of the projection surface and avoid occlusion areas. Finally, according to the data of the ideal rectangular projection area, the corresponding projection parameters, such as projection angle, scaling ratio, offset, etc., are calculated to obtain intelligent projection control parameters.

[0096] Step S4: Process the projection posture control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; evaluate the projection control effect based on the real-time projection image data to obtain projection control optimization effect data.

[0097] In an embodiment of the present invention, according to the intelligent projection control parameters generated in step S3, projection posture control instructions are generated, such as controlling the horizontal and vertical rotation angles of the projector, the focal length of the lens, etc. These instructions are sent to the motor control system of the projector to control the projector to adjust the posture. At the same time, the image sensor collects the projection screen in real time to obtain real-time projection screen data. Using the preset projection reference mark data, the real-time projection screen data is projected with projection reference marks, and key feature points of the projection screen, such as the four corner points of the image, are extracted. The deviations of these feature points from the ideal position are calculated to obtain the geometric distortion. At the same time, the real-time projection screen data is evaluated for image quality, including the evaluation of indicators such as edge clarity, brightness uniformity, and image integrity. The real-time projection screen data is compared with the initial projection screen data to evaluate the color reproduction. According to the preset projection control optimization weights, the geometric distortion, edge clarity, brightness uniformity, image integrity, and color reproduction are weightedly evaluated to obtain the projection control optimization effect data. According to the optimization effect data, the projection control parameters can be further adjusted to achieve the best projection effect. For example, if the edge definition is low, the lens focal length or projection angle can be adjusted; if the brightness uniformity is poor, the light source intensity distribution of the projector can be adjusted; if there is geometric distortion, the projector posture can be adjusted or image correction can be performed according to the distortion. Through iterative optimization, the best projection effect can be achieved.

[0098] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S1 in the flowchart, in this example, step S1 includes:

[0099] Step S11: acquiring target projection image data; the image processor processes the target projection image data into image projection signals, and controls the projection lens assembly to perform image projection, thereby generating initial projection picture data;

[0100] In an embodiment of the present invention, target projection image data is obtained from a preset data source. The data source may be a device such as a computer, a tablet computer, a mobile phone, or a network server or a storage device. The data transmission protocol may adopt a commonly used protocol, such as HDMI, DP, a network protocol, etc. The image data may be in various formats, such as picture formats such as JPEG, PNG, BMP, or video formats such as MP4 and AVI. After receiving the image data, the image processor performs a series of preprocessing operations. These operations include: image format conversion, resolution adjustment, color space conversion (such as RGB to YUV), gamma correction, brightness and contrast adjustment, etc. The purpose of preprocessing is to convert the image data into a format suitable for projector display and optimize the image quality. After the preprocessing is completed, the image data is sent to the projection lens assembly, and by controlling the drive circuit of the projection lens assembly, the focal length, aperture and other parameters of the projection lens are accurately controlled, and finally the initial projection picture is generated on the projection surface.

[0101] Step S12: Based on the preset scanning frequency data, the dual laser radar components are used to perform a left and right dual-path horizontal scanning of the projection area on the initial projection image data to generate left and right projection horizontal scanning data;

[0102] In an embodiment of the present invention, based on a preset scanning frequency (e.g., 10 Hz), the dual laser radar assembly starts to perform a left and right dual-channel horizontal scan of the projection area. Each laser radar is responsible for scanning half of the projection area, thereby improving the scanning efficiency and accuracy. The laser radar calculates the distance of the target point by emitting a laser beam and measuring the time it takes for the laser beam to return. At the same time, the laser radar also records the intensity information of the laser beam, which helps to distinguish the target point from the background noise. The dual laser radars need to be precisely synchronized, which can be achieved using a high-precision clock signal or a dedicated synchronization interface. The left and right laser radars work synchronously and collaboratively, each generating horizontal scanning data. These data contain a large number of three-dimensional points, each of which contains its three-dimensional coordinates (X, Y, Z) and intensity information. The left and right two-way scanning data can provide more comprehensive information about the projection area and reduce the blind spots and errors caused by single-channel scanning. After the data acquisition is completed, the data will be transmitted to the data processing unit for further processing.

[0103] Step S13: using a servo motor to drive the dual laser radar assembly to perform vertical interval layered scanning of the projection area to generate vertical layered scanning data of the projection area;

[0104] In an embodiment of the present invention, a servo motor is used to precisely control the vertical movement of the dual laser radar assembly to achieve vertical layered scanning of the projection area. The servo motor drives the laser radar assembly to scan layer by layer in the vertical direction according to a preset vertical interval (e.g., 5 cm). After each layer of scanning is completed, the servo motor drives the laser radar assembly to move to the next layer and perform the next layer of scanning. This process continues until the vertical scanning of the entire projection area is completed. Vertical layered scanning can obtain the height information of the projection area.

[0105] Step S14: performing point cloud noise filtering on the left and right projection horizontal scanning data and the projection area vertical layered scanning data to obtain projection horizontal scanning point cloud data and projection vertical scanning point cloud data respectively;

[0106] In an embodiment of the present invention, point cloud noise filtering is performed on the left and right projection horizontal scanning data and the projection area vertical layered scanning data obtained in steps S12 and S13. Common noise filtering methods include statistical filtering, spatial filtering, and morphological filtering. Statistical filtering methods, such as those based on median filtering or mean filtering, can effectively remove outliers. Spatial filtering methods, such as filtering methods based on neighborhood search, can remove spatially isolated points. Morphological filtering methods, such as opening operations or closing operations, can remove some small noise points. The selection of a suitable filtering method depends on the characteristics of the point cloud data and the type of noise. When filtering, the filtering parameters need to be carefully adjusted to avoid filtering out valid data. After the filtering is completed, the projection horizontal scanning point cloud data and the projection vertical scanning point cloud data are obtained respectively.

[0107] Step S15: performing point cloud density equalization processing according to the projected horizontal scanning point cloud data and the projected vertical scanning point cloud data, and performing spatiotemporal registration of the projected point cloud to obtain projected scanning point cloud spatial data.

[0108] In an embodiment of the present invention, point cloud density equalization processing is performed on the projected horizontal scanning point cloud data and the projected vertical scanning point cloud data obtained in step S14. Since there are differences in point cloud density in different areas during the laser radar scanning process, density equalization processing is required to ensure the uniformity of the point cloud data and improve the accuracy of subsequent processing. Common point cloud density equalization methods include grid-based interpolation methods and Poisson distribution-based sampling methods. Then, the processed point cloud data is subjected to spatiotemporal registration, and the horizontal scanning data and the vertical scanning data are fused into a complete projected scanning point cloud space data. Spatiotemporal registration needs to consider the time information and spatial information of the point cloud data to ensure the spatial consistency of the point cloud data. Commonly used spatiotemporal registration methods include ICP (iterative closest point) algorithm and feature matching-based registration methods. After the registration is completed, the final projected scanning point cloud space data is obtained.

[0109] Preferably, step S2 comprises the following steps:

[0110] Step S21: performing rough plane fitting of the projection scanning point cloud spatial data to generate initial fitting projection plane parameters;

[0111] Step S22: performing iterative weighted parameter optimization on the initial fitting projection plane parameters to obtain optimized fitting projection plane parameters;

[0112] Step S23: using the projection scanning point cloud spatial data to perform point cloud mapping processing on the optimized fitting projection plane parameters, and extracting the projection surface boundary points to generate projection plane boundary point cloud data;

[0113] Step S24: performing boundary point clustering processing on the projection plane boundary point cloud data, and performing boundary point cluster polygon fitting to obtain projection plane polygon boundary data;

[0114] Step S25: performing projection plane corner point recognition on the projection plane boundary point cloud data to obtain projection plane corner point coordinate data;

[0115] Step S26: Perform projection surface geometric feature processing on the projection plane corner point coordinate data through the projection plane polygon boundary data to generate projection surface geometric feature data.

[0116] In an embodiment of the present invention, a random sampling consensus algorithm (RANSAC) is used to perform rough plane fitting on the projected scanned point cloud spatial data obtained in step S15. The RANSAC algorithm is a robust plane fitting method that can effectively process data sets containing outliers. The algorithm flow is as follows: First, three points are randomly extracted from the point cloud data, and the plane equation determined by these three points is calculated. Then, the distance from all points in the point cloud data to the plane is calculated, and the points with a distance less than a set threshold are taken as inliers, and the remaining points are taken as outliers. If the number of inliers exceeds the set threshold, the plane fitting result is considered valid, and the parameters of the plane (such as the plane normal vector and the distance from the plane to the origin) are recorded. Repeat the above process multiple times, and select the plane with the largest number of inliers as the final fitting result. The iterative weighted least squares method is used to optimize the initial fitting projection plane parameters obtained in step S21. The least squares method is a commonly used parameter estimation method that can minimize the sum of squared errors between the fitting plane and the point cloud data. On the basis of the least squares method, the iterative weighted least squares method introduces a weight factor to assign different weights to different points, thereby improving the fitting accuracy. The algorithm flow is as follows: First, the distance from each point in the point cloud data to the plane is calculated according to the initial fitting plane parameters. Then, different weights are assigned to each point according to the distance size, and the closer the distance, the greater the weight. The new plane parameters are calculated using the weighted least squares method. Repeat the above process many times until the plane parameters converge or reach the preset number of iterations. Using the optimized fitting projection plane parameters obtained in step S22, the projection scanning point cloud space data obtained in step S15 is projected onto the plane. The projection method is to project each three-dimensional point onto the plane to obtain the two-dimensional coordinates of the point on the plane. This can be achieved by the vector projection method. After the projection is completed, the point cloud data projected on the fitting plane is obtained. Next, the boundary points of the projection plane need to be extracted. Edge detection algorithms, such as the Canny operator, can be used to process the projected two-dimensional point cloud data to identify the boundaries of the plane. The projection plane boundary point cloud data generated in step S23 is clustered to divide the boundary points into different clusters. After clustering is completed, polygon fitting is performed on each boundary point cluster. Commonly used polygon fitting methods include the least squares method and the α-shape algorithm. The least squares method can fit simple polygons, while the α-shape algorithm can fit more complex shapes. After fitting, the projection plane polygon boundary data is obtained. Corner point identification is performed on the projection plane boundary point cloud data obtained in step S24. Corner points are points on the boundary with large changes in curvature, and usually represent the corners of the projection plane. Commonly used corner point identification algorithms include the Harris corner point detection algorithm and the FAST corner point detection algorithm. These algorithms identify corner points by calculating local features of image or point cloud data. The selection of a suitable corner point recognition algorithm needs to be adjusted according to the distribution of boundary points and the noise level. The identified corner point coordinates constitute the projection plane corner point coordinate data.Based on the polygon boundary data, the area, perimeter, aspect ratio and other geometric features of the projection surface can be calculated. Based on the corner point coordinate data, the angle, direction and other geometric features of the projection surface can be calculated. These geometric features together describe the shape and size of the projection surface. For example, the area, perimeter and degree of each angle of the polygon can be calculated. The centroid coordinates, major axis and minor axis length of the polygon can also be calculated. These calculations can be implemented through geometric algorithms or mathematical library functions. Finally, the calculated geometric feature data are integrated to generate the projection surface geometric feature data.

[0117] Preferably, step S3 comprises the following steps:

[0118] Step S31: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data;

[0119] Step S32: performing geometric feature analysis of the occlusion area on the projection picture contour data using the projection surface geometric feature data to generate geometric feature data of the occlusion area;

[0120] Step S33: performing projection occlusion spatial distribution analysis on the occlusion area geometric feature data using the projection surface geometric feature data to obtain projection occlusion spatial distribution data;

[0121] Step S34: Determine the ideal projection area according to the projection occlusion spatial distribution data to generate ideal rectangular projection area data; perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters.

[0122] In an embodiment of the present invention, the initial projection picture data generated in step S11 is subjected to contour recognition to obtain the boundary information of the projection picture. The image needs to be preprocessed, such as noise reduction, graying and other operations, to improve the accuracy of contour recognition. Then, an image edge detection algorithm, such as a Canny operator or a Sobel operator, is used to detect edge information in the image. These algorithms identify edges by calculating the gradient or second-order derivative of image pixels. The detected edge information is usually a series of discrete pixel points. In order to obtain a complete contour line, these discrete pixel points need to be connected and fitted. Common methods include curve fitting algorithms, such as the least squares method or the spline interpolation method. In addition, morphological processing methods, such as expansion and corrosion operations, can be used to optimize the shape and continuity of the contour line. Finally, a geometric data representing the contour of the projection picture is obtained, such as a series of coordinate points or polygons, that is, the projection picture contour data. The projection picture contour data is matched and aligned with the projection surface geometric data. This can be achieved by a geometric transformation method, such as an affine transformation or a perspective transformation. After alignment, the occlusion area can be identified by comparing the projection picture contour and the projection surface boundary. If the projection image outline does not overlap with the projection surface boundary in some areas, it is considered that there is occlusion in these areas. Next, it is necessary to perform geometric feature analysis on the occluded area, such as calculating the area, perimeter, centroid coordinates, shape, etc. of the occluded area. These geometric features can be used to describe the size, shape and position of the occluded area. Image segmentation technology can be used to further refine the occluded area and extract more accurate geometric features. Ultimately, all these geometric feature data constitute the geometric feature data of the occluded area. The geometric feature data of the occluded area is combined with the projection surface geometric data to determine the position and range of the occluded area in three-dimensional space. This requires combining the three-dimensional point cloud data obtained in step S1 to project the two-dimensional occluded area information into three-dimensional space. By analyzing the point cloud data, the type, size and position of the occluded object can be determined. Then, according to the distribution of the occluded area in three-dimensional space, the spatial distribution law of the projection occlusion can be analyzed. For example, indicators such as the density, distribution range, and degree of occlusion of the occluded area can be calculated. All these spatial distribution data constitute the spatial distribution data of the projection occlusion. According to the spatial distribution of projection occlusion, a rectangular area that can maximize the use of the projection surface and avoid the occlusion area is determined as the ideal projection area. This can be achieved through optimization algorithms, such as genetic algorithms or simulated annealing algorithms, to find the largest rectangle under the condition of no occlusion. After determining the ideal projection area, it is necessary to calculate the corresponding projection control parameters, such as the horizontal and vertical rotation angles of the projector, the focal length of the lens, the image scaling ratio and offset, etc. These parameters need to be adjusted according to the position and size of the ideal projection area to ensure that the projection image can be accurately projected into the ideal area. These parameters can be calculated based on the projection model and the geometric transformation relationship.Ultimately, all these parameters constitute intelligent projection control parameters, which are used to control the projector to project images.

[0123] Preferably, step S32 includes the following steps:

[0124] Step S321: performing actual projection contour difference analysis on the projection picture contour data through the projection surface geometric feature data to generate actual projection difference contour data;

[0125] Step S322: performing potential occluder point cloud recognition on the projection scanning point cloud space data based on the actual projection difference contour data to obtain potential occluder point cloud data;

[0126] Step S323: clustering the potential occluder point cloud data into occluded area point cloud data to generate occluder clustered spatial area data;

[0127] Step S324: extracting the boundary of the occluder according to the occluder clustering spatial region data to generate occluder region boundary data;

[0128] Step S325: classifying the occlusion region boundary data according to the shape of the occlusion object to obtain occlusion region shape classification data;

[0129] Step S326: performing geometric feature analysis of the occlusion region according to the occlusion region shape classification data and the occlusion region boundary data to generate geometric feature data of the occlusion region.

[0130] In an embodiment of the present invention, the projection surface geometric feature data obtained in step S26 and the projection picture contour data obtained in step S31 are used to perform actual projection contour difference analysis to identify potential occlusion areas. First, the projection picture contour data needs to be projected onto the projection plane determined in step S2. This can be achieved by a geometric transformation method, such as an affine transformation or a perspective transformation, to convert the image coordinates into projection plane coordinates. Then, the projected contour data is compared with the projection surface boundary data extracted in step S2. By calculating the distance difference between the projection picture contour and the projection surface boundary, the areas where the contour and the boundary do not overlap can be identified, and these areas are potential occlusion areas. The distance difference can be expressed as a difference contour, which reflects the degree of deviation between the projection picture contour and the projection surface boundary. The actual projection difference contour data is mapped into a three-dimensional point cloud space. This can be achieved by matching the points on the difference contour with the corresponding point cloud data. The matching method can be selected according to the density and distribution of the point cloud data, such as a nearest neighbor matching method or a feature matching method. After the matching is completed, the three-dimensional point cloud data corresponding to the difference contour can be identified as potential occlusion point cloud data. The selection of a suitable clustering algorithm needs to be adjusted according to the distribution and noise level of the point cloud data. After clustering is completed, each cluster represents a potential occluder area, and the point cloud data it contains constitutes the point cloud information of the area. A boundary extraction method based on convex hull can be used, such as calculating the convex hull of each cluster to obtain the boundary of the cluster. Alternatively, a boundary extraction method based on neighborhood search can be used, such as calculating the neighborhood of each point cloud, and then identifying boundary points based on the distribution of point clouds in the neighborhood. The extracted boundary points constitute the boundary data of the occluder area. For example, the geometric features such as the perimeter, area, and aspect ratio of the boundary data are calculated, and then the shapes are classified into rectangles, circles, triangles, etc. based on these features. Shape classification methods based on machine learning, such as using models such as convolutional neural networks (CNN) to train and classify boundary data, can identify more complex shapes. Finally, the occluded area shape classification data is obtained. According to the shape classification results of the occluder, a suitable geometric feature calculation method is selected. For example, for rectangular occluders, their length, width, area and other geometric features can be calculated; for circular occluders, their radius, area and other geometric features can be calculated; for irregular occluders, their perimeter, area, centroid and other geometric features can be calculated. These geometric feature data can be used to describe the size, shape and position of the occluder, and finally generate the geometric feature data of the occluded area.

[0131] Preferably, step S34 includes the following steps:

[0132] Step S341: Calculate the projection boundary adjustment distance according to the projection occlusion space distribution data to obtain projection boundary adjustment distance data;

[0133] Step S342: performing adjustment boundary mapping on the projection occlusion space distribution data through the projection boundary adjustment distance data to generate target projection adjustment boundary data;

[0134] Step S343: using the target projection adjustment boundary data to identify the non-occluded overlapping area of ​​the occluded area geometric feature data, and performing a projection area visibility evaluation to obtain area visibility evaluation data;

[0135] Step S344: Based on the target projection adjustment boundary data, the projection boundary adjustment distance data is corrected by using the regional visibility evaluation data to generate optimal projection adjustment boundary data;

[0136] Step S345: estimating the projector position offset of the projection surface geometric feature data through the optimal projection adjustment boundary data, and performing gradient descent optimization to generate projector position adjustment data;

[0137] Step S346: determining the ideal projection area according to the projector position adjustment data to generate ideal rectangular projection area data; performing projection image correction control processing on the ideal rectangular projection area data to generate projection image correction control data;

[0138] Step S347: combining the projector position adjustment data and the projection image correction control data into projector intelligent control parameters to obtain intelligent projection control parameters.

[0139] In an embodiment of the present invention, the spatial position and shape of the obstructing object are analyzed to determine the influence range of each obstructing object on the projection area. The influence range can be estimated based on the size of the obstructing object and the distance from the projection surface. Then, according to the preset projection safety margin (for example, a minimum safety distance is set to ensure that the projection image does not overlap with the obstructing object), the distance of the projection boundary that needs to be adjusted is calculated. This distance needs to take into account factors such as the shape and position of the obstructing object and the projection angle of the projector. The calculation method can use a geometric model. For example, for a simple rectangular obstructing object, the distance that the projection boundary needs to move can be directly calculated; for an obstructing object with a complex shape, a more complex geometric calculation method or a numerical simulation method can be used. Finally, the projection boundary adjustment distance data is obtained. The boundary of the projection area is expanded or contracted outward to adjust the projection boundary distance. This mapping process can be understood as a translation transformation of the projection area boundary. The magnitude of the transformation is the distance calculated in step S341. After mapping, a new projection area boundary is generated, that is, the target projection adjustment boundary data. This new boundary excludes the influence area of ​​the obstructing object. The adjusted projection boundary is compared with the geometric feature data of the occluded area to identify the non-overlapping area between the two, that is, the non-occluded overlapping area. Then, the visibility of the non-occluded overlapping area is evaluated. The visibility evaluation can take into account multiple factors, such as light intensity, ambient brightness, brightness of the projector, etc. The visibility score can be calculated by simulating the projection effect or using a model-based evaluation method. If the visibility of some areas is low, the projection boundary distance needs to be further adjusted to improve the visibility of these areas. The correction method can be determined based on the visibility evaluation data and the preset threshold. For example, if the visibility score is lower than the threshold, the projection boundary adjustment distance needs to be increased; otherwise, the projection boundary adjustment distance can be reduced. After correction, the optimal projection adjustment boundary data is generated. By comparing the optimal projection adjustment boundary data and the projection surface geometric feature data, the distance and angle that the projector needs to move can be calculated so that the projected image can be accurately projected into the optimal projection area. This calculation process can use a geometric transformation method, such as an affine transformation or a perspective transformation. In order to further optimize the position of the projector, an optimization algorithm such as a gradient descent algorithm can be used to iteratively adjust the position of the projector until the preset accuracy requirement is reached to obtain the projector position adjustment data. According to the adjusted projector position, the boundary of the projection area is recalculated to determine the ideal rectangular projection area. Then, the projection image correction control processing is performed on the ideal rectangular projection area data. Image correction can correct image deformation caused by projector position adjustment and irregular projection surface shape, such as trapezoidal distortion or perspective distortion. The image correction method can use a geometric transformation method, such as an affine transformation or a perspective transformation, to transform the image to eliminate image deformation.The projector position adjustment data generated in step S345 and the projection image correction control data generated in step S346 are combined to generate the final intelligent projection control parameters. These parameters include the horizontal and vertical rotation angles of the projector, the lens focal length, the image scaling ratio, the offset, and the image correction parameters. These parameters will be sent to the control system of the projector to control the projector to project the image.

[0140] Preferably, step S346 includes the following steps:

[0141] Step S3461: determining the ideal projection area according to the projector position adjustment data, and generating ideal rectangular projection area data;

[0142] Step S3462: performing projection area triangulation on the ideal rectangular projection area data to generate ideal projection sub-area division data;

[0143] Step S3463: constructing a projection sub-region topological relationship for the ideal projection sub-region division data to obtain projection sub-region topological data;

[0144] Step S3464: mapping the projection sub-region corner points according to the projection sub-region topology data, and calculating the local perspective transformation matrix to generate local perspective transformation matrix data;

[0145] Step S3465: Perform light path projection processing on the local perspective transformation matrix data to generate a light path projection angle matrix; perform projection image correction control processing according to the light path projection angle matrix to generate projection image correction control data.

[0146] In an embodiment of the present invention, the projection range of the projection area on the projection surface is calculated according to the adjusted projector position and posture. This calculation process needs to consider factors such as the lens parameters of the projector, the projection distance, and the shape of the projection surface. The calculation can be performed using a geometric projection model or a three-dimensional modeling software. In order to avoid blocking the area, the boundary of the projection area needs to avoid the blocking area identified in step S32. The projection area finally determined is a rectangle, whose position and size can maximize the use of the projection surface and avoid blocking. The four corner point coordinates and length and width information of this rectangular area constitute the ideal rectangular projection area data. The ideal rectangular projection area data generated in step S3461 is triangulated to divide the rectangular area into multiple triangular sub-areas. Triangulation is a commonly used geometric processing method that can divide any polygon into multiple triangles. The result after triangulation is multiple triangles, each triangle is determined by three vertex coordinates. These triangles and their vertex coordinate information constitute the ideal projection sub-area division data. The ideal projection sub-area division data generated in step S3462 is topologically constructed to describe the connection relationship between each triangular sub-area. The topological relationship represents the adjacency relationship between each triangular sub-region and the boundary information of the triangular sub-region. The topological relationship can be represented by an adjacency matrix or an adjacency list. The elements in the adjacency matrix represent whether two triangular sub-regions are adjacent, and the adjacency list stores the list of adjacent sub-regions of each triangular sub-region. To determine the corner point coordinates of each triangular sub-region, these corner point coordinates need to be mapped to the image coordinate system. This mapping process needs to consider the lens parameters and projection transformation relationship of the projector. After the mapping is completed, for each triangular sub-region, its corresponding local perspective transformation matrix can be calculated. The perspective transformation matrix can transform the point coordinates in the triangular sub-region into the image coordinate system. According to the local perspective transformation matrix, the light path projection angle of each triangular sub-region is calculated. The light path projection angle represents the direction of the light from the projector to the projection surface. The calculation of the light path projection angle needs to consider the lens parameters and projection transformation relationship of the projector. Then, these angle information are integrated to generate the light path projection angle matrix. Finally, the projection image is corrected according to the light path projection angle matrix. The correction method can use the inverse perspective transformation method to map the image coordinates to the projection plane coordinate system, thereby eliminating image deformation. Finally, the projection image correction control data is generated.

[0147] Preferably, step S3465 includes the following steps:

[0148] Predicting projection distortion according to the projector position adjustment data to obtain projection distortion prediction data;

[0149] Perform light path projection processing according to the local perspective transformation matrix data to generate a light path projection angle matrix;

[0150] Based on the projection distortion prediction data, the sub-region distortion compensation amount is calculated for the projection sub-region topology data using the light path projection angle matrix to generate the sub-region deformation compensation amount data;

[0151] Remapping the projection pixels according to the sub-region deformation compensation amount data, and performing projection image scaling processing to generate image projection reconstruction data;

[0152] Performing projection area mask processing on the ideal rectangular projection area data through image projection reconstruction data to generate projection area mask data;

[0153] A projection image correction control process is performed based on the projection area mask data and the image projection reconstruction data to generate projection image correction control data.

[0154] In an embodiment of the present invention, a model-based method is used, such as a polynomial model or a radial distortion model to fit the lens distortion characteristics of the projector, and the geometric distortion is calculated according to the position offset of the projector. A data-driven method can also be used, such as using existing projection distortion data to train a machine learning model, and predicting the projection distortion according to the position adjustment data of the projector. The prediction result is a data set describing the degree of projection distortion, which contains the distortion information of each pixel point and constitutes the projection distortion prediction data. For each triangular sub-region, the light direction vector from the projector to each pixel point of the sub-region can be calculated using its corresponding local perspective transformation matrix. The light direction vector can be expressed as a three-dimensional vector, and its direction represents the direction of light propagation. By calculating the angle between the light direction vector and the projection plane, the light projection angle can be obtained. The light projection angle information of all triangular sub-regions is integrated to generate a matrix, each row of the matrix corresponds to a sub-region, each column corresponds to a pixel point, and the element value is the light projection angle corresponding to the pixel point. According to the projection distortion prediction data, the degree of distortion of each pixel point is determined. Then, according to the light path projection angle matrix, the direction of the light corresponding to each pixel point is determined. Combining the projection distortion prediction data and the light direction information, the distortion compensation required for each pixel can be calculated. The compensation can be expressed as a two-dimensional vector, whose direction and size represent the displacement required for the pixel. The distortion compensation of all pixels in each sub-region is integrated to generate a matrix, in which each row corresponds to a sub-region, each column corresponds to a pixel, and the element value is the distortion compensation corresponding to the pixel. According to the deformation compensation data of the sub-region, each pixel is displaced to compensate for the projection distortion. The displacement transformation can be implemented by bilinear interpolation or other interpolation methods. Then, according to the size and shape of the ideal projection area, the projection image is scaled to adapt to the size of the ideal projection area. The scaling can be implemented by bilinear interpolation or other interpolation methods to generate image projection reconstruction data. According to the ideal rectangular projection area data, a binary mask image is generated. In the mask image, the pixel value in the ideal projection area is 1, and the pixel value in other areas is 0. Then, the binary mask image is used to perform mask processing on the image projection reconstruction data, and the pixel values ​​of the areas with pixel values ​​of 0 in the mask image are set to 0, and the pixel values ​​of the areas with pixel values ​​of 1 are retained to generate projection area mask data. The image data after mask processing is used as the final projection image data. The generated projection image correction control data contains the final projection image data and related parameter information, such as image size, pixel position, etc., which is used to control the projector to display the final correction image.

[0155] As an example of the present invention, refer to Figure 3 As shown, Figure 1Detailed implementation steps of step S4 in the embodiment are shown in the flowchart. In this embodiment, step S4 includes:

[0156] Step S41: Processing the intelligent projection control parameters into projection posture control instruction to obtain optimized projection control instruction data;

[0157] In an embodiment of the present invention, the intelligent projection control parameters generated in step S347 are processed to generate instruction data for controlling the posture of the projector. The intelligent projection control parameters include information such as the horizontal and vertical rotation angles, lens focal length, image scaling ratio, and image correction parameters of the projector. These parameters need to be converted into specific motor control instructions to control the various motors of the projector to perform corresponding actions. The conversion process needs to consider the driving characteristics and mechanical structure parameters of each motor. For example, the horizontal rotation angle needs to be converted into the angle or number of pulses of the horizontal motor rotation, the vertical rotation angle needs to be converted into the angle or number of pulses of the vertical motor rotation, and the lens focal length needs to be converted into the control signal of the lens drive motor. After conversion, optimized projection control instruction data is obtained, which includes the control instructions of each motor and is used to control the posture and lens parameters of the projector.

[0158] Step S42: executing the motor control instruction according to the optimized projection control instruction data, and using the image sensor to collect the real-time projection picture to obtain the real-time projection picture data;

[0159] In an embodiment of the present invention, according to the optimized projection control instruction data generated in step S41, each motor of the projector is controlled to perform corresponding actions, and the image sensor is used to collect real-time projection screen data. The motor control instruction data is sent to the motor control system of the projector to drive each motor to rotate or move according to the instruction. At the same time, the image sensor collects the projection screen in real time to obtain real-time projection screen data. The image sensor can be a CMOS or CCD sensor, and its output data is image data, usually a digital image in RGB format.

[0160] Step S43: performing projection reference mark recognition on the real-time projection screen data through the preset projection reference mark data, and extracting key feature points of the projection screen to generate optimized projection screen feature point data;

[0161] In an embodiment of the present invention, the real-time projection screen data collected in step S42 is processed using preset projection reference mark data, the projection reference mark is identified, and the key feature points of the projection screen are extracted. The preset projection reference mark data can be some marks with specific patterns or shapes, such as a checkerboard pattern or a circular mark. Image processing algorithms, such as template matching algorithms or corner point detection algorithms, are used to identify the positions of these marks in the real-time screen. After the reference marks are identified, the key feature points of the projection screen, such as the four corner points or other feature points of the image, can be extracted according to the positions and geometric relationships of the marks to generate optimized projection screen feature point data.

[0162] Step S44: Calculate the geometric distortion of the optimized projection picture feature point data to generate optimized geometric distortion data; evaluate the edge clarity of the real-time projection picture data to obtain edge clarity data; evaluate the brightness uniformity of the real-time projection picture data to obtain brightness uniformity data; evaluate the image projection integrity of the real-time projection picture data to obtain image projection integrity data;

[0163] In an embodiment of the present invention, the geometric distortion of the projected image is calculated based on the optimized projection picture feature point data. The geometric distortion reflects the degree of deviation between the projected image and the ideal image. The calculation method can be calculated based on the position of the feature point and the preset ideal position. Then, the edge clarity evaluation, brightness uniformity evaluation and image projection integrity evaluation are performed on the real-time projection picture data. The edge clarity evaluation can use an edge detection algorithm, such as the Sobel operator or the Canny operator, to calculate the clarity of the image edge. The brightness uniformity evaluation can calculate the brightness difference in different areas of the image. The image projection integrity evaluation can determine whether the projected image is complete, whether there is any missing or deformation. Finally, the optimized geometric distortion data, edge clarity data, brightness uniformity data and image projection integrity data are obtained.

[0164] Step S45: performing color restoration evaluation on the real-time projection picture data and the initial projection picture data based on the optimized projection picture feature point data to obtain color restoration data;

[0165] In an embodiment of the present invention, the real-time projection picture data is aligned with the initial projection picture data, which can be achieved by the key feature points extracted in step S43. After alignment, the color information of the two images can be compared to calculate the color restoration degree. The color restoration degree reflects the degree of color consistency between the projected image and the original image. The calculation method can use a color space distance metric, such as a CIE76 color space distance or a CIE94 color space distance, to obtain color restoration degree data.

[0166] Step S46: using the preset projection control optimization weight data to perform a weighted evaluation of the control effect on the optimized geometric distortion data, edge clarity data, brightness uniformity data, image projection integrity data and color restoration data, and obtaining projection control optimization effect data.

[0167] In an embodiment of the present invention, the preset projection control optimization weight data is used to perform weighted evaluation on the optimized geometric distortion data, edge clarity data, brightness uniformity data, image projection integrity data, and color reproduction data obtained in step S44 and step S45 to obtain projection control optimization effect data. The preset projection control optimization weight data is set according to actual needs. For example, the weights of different indicators can be adjusted according to application scenarios and user preferences. The weighted evaluation method can adopt linear weighting or nonlinear weighting and other methods. The linear weighting method is to multiply the scores of each indicator by the corresponding weight and then sum them; the nonlinear weighting method can adopt different weight calculation methods according to different combinations of indicators. The projection control optimization effect data finally obtained reflects the overall performance of the projection system. This data can be used as feedback information to adjust the projector parameters to achieve the best projection effect.

[0168] Preferably, the present invention also provides a projector, referring to Figure 4 As shown, the projector includes a projection host and a dual laser radar assembly, the projection host is provided with a projection lens assembly, an image processor and a main controller, the dual laser radar assembly includes a first laser radar, a second laser radar and two servo motors, the first laser radar and the second laser radar are respectively fixed on both sides of the projection host, and are driven by the servo motors to perform vertical scanning, the dual laser radar assembly, the projection lens assembly and the image processor are electrically connected to the main controller through a data bus, wherein the main controller is used to execute the projector control method as described above, and the main controller includes the following modules:

[0169] The projection point cloud processing module is used to obtain the target projection image data; the image processor processes the target projection image data with image projection signals, and controls the projection lens assembly to project the image, generating initial projection screen data; the dual laser radar assembly is used to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining the projection scanning point cloud spatial data;

[0170] The projection surface feature extraction module is used to extract the projection surface boundary points from the projection scanning point cloud space data to generate the projection plane boundary point cloud data; perform projection surface geometric feature processing based on the projection plane boundary point cloud data to generate the projection surface geometric feature data;

[0171] The projection control parameter optimization module is used to identify the projection picture contour of the initial projection picture data to obtain the projection picture contour data; to analyze the geometric characteristics of the occlusion area of ​​the projection picture contour data through the projection surface geometric characteristic data to generate the geometric characteristic data of the occlusion area; to determine the ideal projection area according to the geometric characteristic data of the occlusion area to generate the ideal rectangular projection area data; to perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain the intelligent projection control parameters;

[0172] The projection attitude control module is used to process the projection attitude control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; the projection control effect is evaluated based on the real-time projection image data to obtain projection control optimization effect data.

[0173] The present application is to use a dual laser radar component to perform precise spatial scanning and point cloud registration of the projection area. The present invention can intelligently identify and adapt to different positions, shapes and sizes of the projection surface, realize automatic adjustment and optimization of the projection effect, and ensure that the best projection effect can be achieved in various environments. Through precise image projection signal processing and the generation of the initial projection picture, it is ensured that the projection system has a highly consistent projection basis in different environments. Based on the left and right horizontal scanning of the dual laser radar at a preset scanning frequency, the projection area is effectively divided into left and right scanning channels, so that the details of the horizontal area of ​​the projection surface can be fully captured. The radar component is then controlled by a servo motor to realize vertical interval layered scanning, which further refines the point cloud distribution in the vertical direction and ensures the full capture of the depth information of the projection surface. This combination of horizontal and vertical bidirectional scanning makes the spatial feature recognition of the projection area more accurate and avoids the loss of information caused by complex environments or obstacles. The initial geometric features of the projection area are effectively obtained by coarse plane fitting of the projection scanning point cloud spatial data. By iteratively weighting and optimizing parameters, the accuracy of projection plane fitting is further improved, so that the projection plane parameters can fit the actual environment more accurately, especially when facing irregular or inclined projection surfaces, ensuring the flatness and integrity of the projection image. Subsequently, point cloud mapping processing and boundary point extraction accurately identify the boundary contour of the projection surface, greatly improving the adaptive ability to complex boundary shapes. By intelligently analyzing the geometric feature data of the occluded area, the optimal projection area can be flexibly determined, thereby avoiding the phenomenon of image loss caused by obstacles and ensuring the consistency of the viewing experience. The generated projection control parameters are combined with real-time image acquisition and feedback evaluation, so that the projector can intelligently adjust the projection posture according to environmental changes and dynamically optimize the display effect, so that the projector can maintain a stable correction effect in different environments without manual adjustment, significantly improving the projection accuracy and automatic adaptability of the projection effect.

[0174] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0175] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A projector control method, characterized in that: Applied to a projector, the projector includes a projection host and a dual laser radar component, wherein the projection host includes a projection lens component, an image processor and a main controller, the dual laser radar component, the projection lens component and the image processor are electrically connected to the main controller via a data bus, and the projector control method includes the following steps: Step S1: acquiring target projection image data; the image processor processes the target projection image data for image projection signals, and controls the projection lens assembly to perform image projection, generating initial projection screen data; using the dual laser radar assembly to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining projection scanning point cloud spatial data; Step S2: extracting projection surface boundary points from the projection scanning point cloud spatial data to generate projection plane boundary point cloud data; processing projection surface geometric features according to the projection plane boundary point cloud data to generate projection surface geometric feature data; Step S3: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data; performing occlusion area geometric feature analysis on the projection picture contour data through the projection surface geometric feature data to generate occlusion area geometric feature data; determining an ideal projection area according to the occlusion area geometric feature data to generate ideal rectangular projection area data; performing intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters; Step S4: Process the projection posture control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; evaluate the projection control effect based on the real-time projection image data to obtain projection control optimization effect data.

2. The projector control method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: acquiring target projection image data; the image processor processes the target projection image data into image projection signals, and controls the projection lens assembly to perform image projection, thereby generating initial projection picture data; Step S12: Based on the preset scanning frequency data, the dual laser radar components are used to perform a left and right dual-path horizontal scanning of the projection area on the initial projection image data to generate left and right projection horizontal scanning data; Step S13: using a servo motor to drive the dual laser radar assembly to perform vertical interval layered scanning of the projection area to generate vertical layered scanning data of the projection area; Step S14: performing point cloud noise filtering on the left and right projection horizontal scanning data and the projection area vertical layered scanning data to obtain projection horizontal scanning point cloud data and projection vertical scanning point cloud data respectively; Step S15: performing point cloud density equalization processing according to the projected horizontal scanning point cloud data and the projected vertical scanning point cloud data, and performing spatiotemporal registration of the projected point cloud to obtain projected scanning point cloud spatial data.

3. The projector control method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing rough plane fitting of the projection scanning point cloud spatial data to generate initial fitting projection plane parameters; Step S22: performing iterative weighted parameter optimization on the initial fitting projection plane parameters to obtain optimized fitting projection plane parameters; Step S23: using the projection scanning point cloud spatial data to perform point cloud mapping processing on the optimized fitting projection plane parameters, and extracting the projection surface boundary points to generate projection plane boundary point cloud data; Step S24: performing boundary point clustering processing on the projection plane boundary point cloud data, and performing boundary point cluster polygon fitting to obtain projection plane polygon boundary data; Step S25: performing projection plane corner point recognition on the projection plane boundary point cloud data to obtain projection plane corner point coordinate data; Step S26: Perform projection surface geometric feature processing on the projection plane corner point coordinate data through the projection plane polygon boundary data to generate projection surface geometric feature data.

4. The projector control method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing projection picture contour recognition on the initial projection picture data to obtain projection picture contour data; Step S32: performing geometric feature analysis of the occlusion area on the projection picture contour data using the projection surface geometric feature data to generate geometric feature data of the occlusion area; Step S33: performing projection occlusion spatial distribution analysis on the occlusion area geometric feature data using the projection surface geometric feature data to obtain projection occlusion spatial distribution data; Step S34: Determine the ideal projection area according to the projection occlusion spatial distribution data to generate ideal rectangular projection area data; perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain intelligent projection control parameters.

5. The projector control method according to claim 4, characterized in that: Step S32 includes the following steps: Step S321: performing actual projection contour difference analysis on the projection picture contour data through the projection surface geometric feature data to generate actual projection difference contour data; Step S322: performing potential occluder point cloud recognition on the projection scanning point cloud space data based on the actual projection difference contour data to obtain potential occluder point cloud data; Step S323: clustering the potential occluder point cloud data into occluded area point cloud data to generate occluder clustered spatial area data; Step S324: extracting the boundary of the occluder according to the occluder clustering spatial region data to generate occluder region boundary data; Step S325: classifying the occlusion region boundary data according to the shape of the occlusion object to obtain occlusion region shape classification data; Step S326: performing geometric feature analysis of the occlusion region according to the occlusion region shape classification data and the occlusion region boundary data to generate geometric feature data of the occlusion region.

6. The projector control method according to claim 4, characterized in that: Step S34 includes the following steps: Step S341: Calculate the projection boundary adjustment distance according to the projection occlusion space distribution data to obtain projection boundary adjustment distance data; Step S342: performing adjustment boundary mapping on the projection occlusion space distribution data through the projection boundary adjustment distance data to generate target projection adjustment boundary data; Step S343: using the target projection adjustment boundary data to identify the non-occluded overlapping area of ​​the occluded area geometric feature data, and performing a projection area visibility evaluation to obtain area visibility evaluation data; Step S344: Based on the target projection adjustment boundary data, the projection boundary adjustment distance data is corrected by using the regional visibility evaluation data to generate optimal projection adjustment boundary data; Step S345: estimating the projector position offset of the projection surface geometric feature data through the optimal projection adjustment boundary data, and performing gradient descent optimization to generate projector position adjustment data; Step S346: determining the ideal projection area according to the projector position adjustment data to generate ideal rectangular projection area data; performing projection image correction control processing on the ideal rectangular projection area data to generate projection image correction control data; Step S347: combining the projector position adjustment data and the projection image correction control data into projector intelligent control parameters to obtain intelligent projection control parameters.

7. The projector control method according to claim 6, characterized in that: Step S346 includes the following steps: Step S3461: determining the ideal projection area according to the projector position adjustment data, and generating ideal rectangular projection area data; Step S3462: performing projection area triangulation on the ideal rectangular projection area data to generate ideal projection sub-area division data; Step S3463: constructing a projection sub-region topological relationship for the ideal projection sub-region division data to obtain projection sub-region topological data; Step S3464: mapping the projection sub-region corner points according to the projection sub-region topology data, and calculating the local perspective transformation matrix to generate local perspective transformation matrix data; Step S3465: Perform light path projection processing on the local perspective transformation matrix data to generate a light path projection angle matrix; perform projection image correction control processing according to the light path projection angle matrix to generate projection image correction control data.

8. The projector control method according to claim 7, characterized in that: Step S3465 includes the following steps: Predicting projection distortion according to the projector position adjustment data to obtain projection distortion prediction data; Perform light path projection processing according to the local perspective transformation matrix data to generate a light path projection angle matrix; Based on the projection distortion prediction data, the sub-region distortion compensation amount is calculated for the projection sub-region topology data using the light path projection angle matrix to generate the sub-region deformation compensation amount data; Remapping the projection pixels according to the sub-region deformation compensation amount data, and performing projection image scaling processing to generate image projection reconstruction data; Performing projection area mask processing on the ideal rectangular projection area data through image projection reconstruction data to generate projection area mask data; A projection image correction control process is performed based on the projection area mask data and the image projection reconstruction data to generate projection image correction control data.

9. The projector control method according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: Processing the intelligent projection control parameters into projection posture control instruction to obtain optimized projection control instruction data; Step S42: executing the motor control instruction according to the optimized projection control instruction data, and using the image sensor to collect the real-time projection picture to obtain the real-time projection picture data; Step S43: performing projection reference mark recognition on the real-time projection screen data through the preset projection reference mark data, and extracting key feature points of the projection screen to generate optimized projection screen feature point data; Step S44: Calculate the geometric distortion of the optimized projection picture feature point data to generate optimized geometric distortion data; evaluate the edge clarity of the real-time projection picture data to obtain edge clarity data; evaluate the brightness uniformity of the real-time projection picture data to obtain brightness uniformity data; evaluate the image projection integrity of the real-time projection picture data to obtain image projection integrity data; Step S45: evaluating the color restoration degree of the real-time projection picture data and the initial projection picture data based on the optimized projection picture feature point data to obtain color restoration degree data; Step S46: using the preset projection control optimization weight data to perform a weighted evaluation of the control effect on the optimized geometric distortion data, edge clarity data, brightness uniformity data, image projection integrity data and color restoration data, and obtaining projection control optimization effect data.

10. A projector, characterized in that: The projector includes a projection host and a dual laser radar component. The projection host is provided with a projection lens component, an image processor and a main controller. The dual laser radar component includes a first laser radar, a second laser radar and two servo motors. The first laser radar and the second laser radar are respectively fixed on both sides of the projection host and driven by the servo motors to perform vertical scanning. The dual laser radar component, the projection lens component and the image processor are electrically connected to the main controller through a data bus. The main controller is used to execute the projector control method according to claim 1. The main controller includes the following modules: The projection point cloud processing module is used to obtain the target projection image data; the image processor processes the target projection image data with image projection signals, and controls the projection lens assembly to project the image, generating initial projection screen data; the dual laser radar assembly is used to perform spatiotemporal registration of the projection point cloud on the initial projection screen data, obtaining the projection scanning point cloud spatial data; The projection surface feature extraction module is used to extract the projection surface boundary points from the projection scanning point cloud space data to generate the projection plane boundary point cloud data; perform projection surface geometric feature processing based on the projection plane boundary point cloud data to generate the projection surface geometric feature data; The projection control parameter optimization module is used to identify the projection picture contour of the initial projection picture data to obtain the projection picture contour data; to analyze the geometric characteristics of the occlusion area of ​​the projection picture contour data through the projection surface geometric characteristic data to generate the geometric characteristic data of the occlusion area; to determine the ideal projection area according to the geometric characteristic data of the occlusion area to generate the ideal rectangular projection area data; to perform intelligent projection control parameter processing on the ideal rectangular projection area data to obtain the intelligent projection control parameters; The projection attitude control module is used to process the projection attitude control instructions for the intelligent projection control parameters, and use the image sensor to collect real-time projection images to obtain real-time projection image data; the projection control effect is evaluated based on the real-time projection image data to obtain projection control optimization effect data.

Images