Wall sound insulation material laying auxiliary positioning system based on wall surface image

Abstract

The present application relates to the technical field of image recognition, in particular to a wall sound insulation material laying auxiliary positioning system based on wall surface image. In the present application, wall surface pixel point color data covering the wall surface to be laid, wall corner area, door and window edge line area and positioning sticker pattern area are acquired, and wall surface reference positioning points are determined based on positioning sticker recognition; further, wall sound insulation material vertical reference line, wall surface bottom boundary horizontal straight line, laying origin point coordinates, laying baseline coordinates and cutting node coordinates are determined by combining wall surface image gray value, wall surface edge pixel point coordinates and wall surface physical structure straight line data set; finally, wall sound insulation material laying positioning map is constructed, realizing continuous positioning processing from wall surface image acquisition, positioning reference extraction, wall surface structure recognition to laying path generation and cutting position determination, so as to provide intuitive and unified coordinate construction reference for vertical strip board piecewise splicing and laying process.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, and in particular to an auxiliary positioning system for laying sound insulation materials on walls based on wall images. Background Technology

[0002] A wall sound insulation material laying auxiliary positioning system refers to a system used to assist in positioning the sound insulation material during wall construction. It uses measuring devices to mark the wall surface to guide the laying process. This type of system usually involves dividing and positioning the wall area and indicating the material laying position. It often combines image acquisition equipment, measuring tools, and basic image processing methods to acquire and analyze wall information, so as to achieve positional reference and auxiliary guidance during the sound insulation material laying process.

[0003] Traditional methods for assisted positioning of wall sound insulation materials, while able to use measuring devices, image acquisition equipment, and basic image processing techniques to mark the location and provide a reference for laying, suffer from several drawbacks when there are corner areas, door and window edges, or multiple sheets of sound insulation material being laid continuously. These issues include insufficient detail in the division of wall areas, unstable positioning benchmarks, unclear cutting points at door and window junctions, and difficulty in synchronizing the laying lines with the actual wall structure. Consequently, the construction process often involves repeated checks, repeated layouts, and numerous on-site corrections, resulting in insufficient positioning continuity, unintuitive cutting judgments, and weak consistency in construction references. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and propose an auxiliary positioning system for laying sound insulation materials based on wall images.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a wall sound insulation material laying auxiliary positioning system based on wall images includes:

[0006] The wall image acquisition module acquires the color data of wall pixels covering the wall to be covered, the corner area of ​​the wall to be covered, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall.

[0007] The positioning sticker recognition module determines the coordinates of candidate positioning points on the wall based on the color data of the pixels on the wall and the pattern area of ​​the positioning sticker, and then filters the baseline positioning points on the wall based on the coordinates of the candidate positioning points.

[0008] The wall structure extraction module determines the grayscale value of the wall image and the coordinates of the wall edge pixels by referring to the wall pixel color data, and summarizes the straight line dataset of the wall physical structure based on the wall edge pixel coordinates.

[0009] The baseline determination module determines the vertical reference line of the wall sound insulation material based on the wall reference positioning point, and selects the horizontal straight line of the bottom boundary of the wall based on the straight line dataset of the wall physical structure. Based on the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall, the coordinates of the starting point of the wall sound insulation material and the coordinates of the wall sound insulation material laying baseline are set.

[0010] The positioning map generation module determines the coordinates of the wall sound insulation material cutting nodes based on the straight line dataset of the wall physical structure and the baseline coordinates of the wall sound insulation material laying. Based on the coordinates of the starting point coordinates of the wall sound insulation material, the baseline coordinate data of the wall sound insulation material laying, and the coordinates of the wall sound insulation material cutting nodes, it constructs a positioning map of the wall sound insulation material laying.

[0011] As a further aspect of the present invention, the wall image acquisition module includes:

[0012] The color extraction submodule acquires a colored wall image covering the wall to be paved, the corner area corresponding to the wall to be paved, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall. It scans the wall pixels in the colored wall image line by line and extracts the red channel color value, green channel color value, and blue channel color value corresponding to each wall pixel to obtain a three-channel color value group of the wall pixel.

[0013] The region alignment submodule establishes a correspondence between the pixel coordinates and three-channel color values ​​of the wall surface, the corner area corresponding to the wall surface, the door and window edge area, and the positioning sticker pattern area based on the three-channel color value group of the wall surface pixels, and obtains the region color correspondence set.

[0014] The color combination submodule extracts the red, green, and blue channel color values ​​corresponding to the pixel coordinates of each region in the region color correspondence set. Based on the horizontal and vertical arrangement of the pixels, it establishes a wall image coordinate system with the top left pixel as the origin, the horizontal direction as the horizontal coordinate direction, and the vertical direction as the vertical coordinate direction, and generates wall pixel color data.

[0015] As a further aspect of the present invention, the positioning sticker recognition module includes:

[0016] The sticker screening module selects the coordinates of the marker pixels whose color ratios match the preset color ratio range of the marker pixels based on the red channel color values, green channel color values, and blue channel color values ​​of the marker pixel pattern area in the wall pixel color data. It merges multiple sets of adjacent marker pixel coordinates and calculates the geometric center coordinates of each set of marker pixel coordinates as the coordinates of the candidate positioning points on the wall.

[0017] The parameter acquisition submodule establishes the point correspondence between the coordinates of the candidate positioning points on the wall and the coordinates of the standard feature points of the corresponding pre-set positioning sticker pattern. Through a random sampling consensus algorithm, it iteratively extracts a portion of the corresponding point pairs in the point correspondence, and obtains the perspective transformation parameters from the coordinates of the standard feature points of the positioning sticker pattern to the coordinates of the candidate positioning points on the wall in each iteration of the extracted portion of the corresponding point pairs, thus obtaining the perspective mapping parameter set.

[0018] The interior point retention submodule, based on the perspective mapping parameter group, maps the coordinates of the standard feature points of the positioning sticker pattern to the wall image coordinate system to obtain the theoretical projection coordinates of the wall. It calculates the two-dimensional plane straight-line distance between the coordinates of the corresponding candidate positioning points on the wall and the theoretical projection coordinates of the wall for each iteration, retains the coordinates of the candidate positioning points on the wall corresponding to a single iteration that are less than a preset distance judgment threshold and counts the number. It selects the set of candidate positioning point coordinates on the wall with the largest number as the wall reference positioning point.

[0019] As a further aspect of the present invention, the wall structure extraction module includes:

[0020] The grayscale conversion module obtains the brightness-weighted average of the red, green, and blue channel color values ​​of the wall to be paved, the corner area corresponding to the wall to be paved, and the door and window edge area in the wall pixel color data, and obtains the grayscale value of the wall image of each wall pixel.

[0021] The edge selection submodule calculates the horizontal and vertical brightness differences between the grayscale value of each wall pixel and the corresponding grayscale values ​​of its horizontal and vertical adjacent wall pixels. Based on the horizontal and vertical brightness differences, it determines the brightness variation range among multiple wall pixels and selects the coordinates of wall edge pixels whose brightness variation range exceeds a preset edge determination threshold.

[0022] The straight line conversion submodule calculates the polar angle and polar radius values ​​of the wall edge pixel coordinates in the preset polar coordinate system using the Hough transform algorithm, counts the cumulative number of parameter votes, selects the target polar angle and polar radius value combination that is higher than the preset straight line voting threshold, converts it into a straight line of the wall in the wall image coordinate system, and generates a straight line dataset of the wall physical structure.

[0023] As a further aspect of the present invention, the laying baseline determination module includes:

[0024] The direction fitting submodule combines the arrangement direction of the wall reference positioning points and the standard feature points of the positioning sticker pattern to fit the horizontal direction vector and the vertical direction vector of the wall, obtain the direction vector of the wall line in the wall physical structure straight line data, and calculate the cosine value of the angle between the straight line direction vector and the vertical direction vector of the wall.

[0025] The baseline screening module selects wall straight lines whose included angle cosine value is greater than a preset parallelism threshold as vertical reference lines for wall sound insulation materials. At the same time, it selects wall straight lines from the wall physical structure straight line dataset that are parallel to the horizontal vector of the wall and located at the lower boundary of the corner area corresponding to the wall to be laid as horizontal straight lines at the bottom boundary of the wall.

[0026] The coordinate translation submodule sets the coordinates of the intersection point of the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall as the origin coordinates of the wall sound insulation material. At the same time, starting from the origin coordinates of the wall sound insulation material, it translates at equal distances along the horizontal vector direction of the wall according to the preset standard width value of the wall sound insulation material to determine the baseline coordinates of the wall sound insulation material.

[0027] As a further aspect of the present invention, the positioning map generation module includes:

[0028] The node determination submodule, referring to the wall physical structure straight line dataset, filters the wall straight lines located in the door and window edge line area, and determines the door and window boundary straight lines according to the extension direction and position continuity of the wall straight lines in the door and window edge line area. Based on the door and window boundary straight lines and the baseline coordinates of the wall sound insulation material laying, it determines the coordinates of the two-dimensional plane intersection point, which is used as the coordinates of the wall sound insulation material cutting node.

[0029] The scaling conversion submodule calls the screen display size scaling factor to adjust the positional relationship between the starting point coordinates of the wall sound insulation material, the laying baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, and calculates the display scaling coordinate group.

[0030] The image construction submodule, based on the display scaling coordinate group, splices and reassembles the display scaling coordinates corresponding to the starting point coordinates of the wall sound insulation material, the baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, to generate a positioning map of the wall sound insulation material based on the wall image.

[0031] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0032] In this invention, by acquiring the color data of wall pixels covering the wall surface to be laid, the corner area, the door and window edge area, and the area of ​​the positioning sticker pattern, and determining the wall reference positioning point based on the positioning sticker identification, and further combining the wall image grayscale value, the wall edge pixel coordinates, and the wall physical structure straight line dataset, the vertical reference line of the wall sound insulation material, the horizontal straight line of the bottom boundary of the wall, the coordinates of the starting point, the coordinates of the laying baseline, and the coordinates of the cutting node are determined. Finally, a wall sound insulation material laying positioning map is constructed. This realizes continuous positioning processing from wall image acquisition, positioning reference extraction, wall structure recognition to laying path generation and cutting position determination. As a result, it can provide an intuitive and unified coordinate construction reference for the vertical strip splicing laying process, which can improve the laying positioning accuracy, enhance the clarity of cutting judgment in the door and window junction area, reduce repeated measurement and repetitive line laying operations, and improve the construction connection and the completeness of the positioning map output. Attached Figure Description

[0033] Figure 1 This is a flowchart of the system modules of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0035] Please see Figure 1 The wall image-based auxiliary positioning system for laying sound insulation materials includes:

[0036] The wall image acquisition module acquires the color data of wall pixels covering the wall to be covered, the corner area of ​​the wall to be covered, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall.

[0037] The positioning sticker recognition module determines the coordinates of candidate positioning points on the wall based on the color data of the pixels on the wall and the pattern area of ​​the positioning sticker, and then filters the baseline positioning points on the wall based on the coordinates of the candidate positioning points.

[0038] The wall structure extraction module determines the grayscale value of the wall image and the coordinates of the wall edge pixels by referring to the wall pixel color data, and summarizes the straight line dataset of the wall physical structure based on the wall edge pixel coordinates.

[0039] The baseline determination module determines the vertical reference line of the wall sound insulation material based on the wall reference positioning point, and selects the horizontal straight line of the bottom boundary of the wall based on the straight line dataset of the wall physical structure. Based on the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall, the coordinates of the starting point of the wall sound insulation material and the coordinates of the wall sound insulation material laying baseline are set.

[0040] The positioning map generation module determines the coordinates of the wall sound insulation material cutting nodes based on the straight line dataset of the wall physical structure and the baseline coordinates of the wall sound insulation material laying. Based on the coordinates of the starting point coordinates of the wall sound insulation material, the baseline coordinate data of the wall sound insulation material laying, and the coordinates of the wall sound insulation material cutting nodes, it constructs a positioning map of the wall sound insulation material laying.

[0041] The wall image acquisition module includes:

[0042] The color extraction submodule acquires a colored wall image covering the wall to be paved, the corner area corresponding to the wall to be paved, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall. It scans the wall pixels in the colored wall image line by line and extracts the red channel color value, green channel color value, and blue channel color value corresponding to each wall pixel to obtain a three-channel color value group of the wall pixel.

[0043] Firstly, in the scenario of auxiliary positioning for laying wall sound insulation materials, there are various methods for laying wall sound insulation materials, including laying the entire board at once, horizontally segmented splicing, and vertically splicing strips one by one. This involves a complete processing chain combining the wall's reference positioning point, the horizontal straight line of the wall's bottom boundary, the laying baseline coordinates obtained by equidistant translation according to standard width, and the intersection of the door / window boundary line and the laying baseline as the cutting node. This chain is suitable for laying wall sound insulation materials vertically spliced ​​strips one by one. The construction workers first affix positioning stickers to the visible area of ​​the wall to be covered, then use a handheld industrial camera, tablet terminal camera, or acquisition device with a fixed-focus lens to capture images of the wall one or more times. The captured images simultaneously cover the wall to be covered, the corner area, the door / window edge area, and the area of ​​the positioning sticker pattern. The image data is then imported into MATLAB, and MATLAB... The image storage order reads pixel data row by row, starting from row 1. For each wall pixel read, the red, green, and blue channel values ​​are retrieved from that pixel's data address. These three values ​​are recorded in the range of 0 to 255. When a channel value is between 0 and 85, it is recorded as a low channel value; 0 indicates no color response for that channel, and 85 corresponds to approximately 1 / 3 of 255. Values ​​between 86 and 170 are recorded as medium channel values, placing them in the middle of the range. Values ​​between 171 and 255 are recorded as high channel values, with values ​​above 171 approaching the upper range. Then, the red, green, and blue channel values ​​of the same pixel are grouped into a set of three-channel color values ​​for the wall pixel in a fixed order, and this set of data, along with the row and column number of the pixel, is synchronously written to the cache table.

[0044] The region alignment submodule establishes the correspondence between the pixel coordinates and three-channel color values ​​of the wall surface, the corner area, the door and window edge area, and the positioning sticker pattern area based on the three-channel color value group of the wall surface pixels, and obtains the region color correspondence set.

[0045] The process involves retrieving pixel coordinates and three-channel color values ​​from the cache table to create categorized records for the wall surface to be covered, corner areas, door and window edges, and positioning sticker patterns, point by point within the same coordinate framework. During execution, the horizontal and vertical coordinates of each pixel are first read, then combined to form a position index. The corresponding red, green, and blue values ​​are combined to form a color index, and both the position and color indices are written to the judgment queue. Subsequently, the process judges each pixel point according to its actual boundary position on the construction screen. Pixel coordinates falling within the closed area of ​​the main wall outline are written to the wall surface record; pixel coordinates falling within the corner transition zone are written to the wall surface record. Write the corner area record; write the door / window edge area record when the pixel coordinates fall within the preset width range on both sides of the door or window frame line; write the positioning sticker pattern area record when the pixel coordinates fall within the closed outline of the positioning sticker pattern. When the same pixel meets multiple area conditions simultaneously, retain the write result once in the order of positioning sticker pattern area, door / window edge area, corner area, and wall area to be covered, and delete all subsequent category results. If the same pixel does not meet any area condition, do not write the area color correspondence set. After traversing all pixels, obtain the correspondence between pixel coordinates and three-channel color values.

[0046] The color combination submodule extracts the red, green, and blue channel color values ​​corresponding to the pixel coordinates of each region in the region color correspondence set. It then establishes a wall image coordinate system with the top-left pixel as the origin, the horizontal axis as the horizontal axis, and the vertical axis as the vertical axis, according to the horizontal and vertical arrangement of the pixels, and generates wall pixel color data.

[0047] The top-left pixel is fixed as the origin, the rightward arrangement direction is fixed as the horizontal coordinate increasing direction, and the downward arrangement direction is fixed as the vertical coordinate increasing direction. Then, the red, green, and blue channel values ​​corresponding to the pixel coordinates of each region are extracted sequentially. Next, the pixels are sorted first by vertical coordinate from smallest to largest, and then by horizontal coordinate from smallest to largest among the records with the same vertical coordinate. After sorting, the horizontal coordinate, vertical coordinate, red channel value, green channel value, and blue channel value are concatenated into a single pixel record. If the vertical coordinates of two adjacent records are different, it is determined that a newline has been written. If the vertical coordinates of two adjacent records are the same and the difference in horizontal coordinates is 1, they are determined to be consecutive pixels in the same row. If the vertical coordinates of two adjacent records are the same and the difference in horizontal coordinates is greater than 1, it is determined that there is a gap in the same row, and an empty coordinate marker is inserted at the gap position. After all records are completed, the wall pixel color data is formed for both the positioning sticker recognition and wall structure extraction.

[0048] The location sticker recognition module includes:

[0049] The sticker screening module selects the coordinates of the marker pixels whose color ratios match the preset color ratio range of the marker pixels based on the red, green and blue channel color values ​​of the marker pixel pattern area in the wall pixel color data. It merges multiple sets of adjacent marker pixel coordinates and calculates the geometric center coordinates of each set of marker pixel coordinates as the coordinates of the candidate positioning points on the wall.

[0050] The color ratio range of the positioning sticker includes the proportion range of the red channel, the proportion range of the green channel, and the proportion range of the blue channel;

[0051] Within the positioning sticker pattern area, pixel filtering is performed. The red, green, and blue channel values ​​of each pixel in the pattern area are read point by point, and then the sum of the three channels is calculated: Sum of three channels = Red channel value + Green channel value + Blue channel value. When the sum of the three channels = 0, the pixel is directly deleted and not included in subsequent proportion determination. When the sum of the three channels > 0, the proportions of the red, green, and blue channels are calculated: Red channel proportion = Red channel value / Sum of three channels; Green channel proportion = Green channel value / Sum of three channels; Blue channel proportion = Blue channel value / Sum of three channels. Then, the three proportions are compared with the preset positioning sticker color proportion range. If the red channel proportion is below the lower limit of the red proportion, the pixel is deleted; if the red channel proportion is above the upper limit of the red proportion, the pixel is deleted; if the green channel proportion is below the lower limit of the green proportion, the pixel is deleted; if the green channel proportion is above the upper limit of the green proportion, the pixel is deleted; if the blue channel proportion is below the blue channel proportion, the pixel is deleted. The pixel is deleted when the percentage falls below the lower limit; the pixel is deleted when the percentage of the blue channel is higher than the upper limit of the blue channel percentage; the pixel coordinates are retained as the coordinates of the identifier pixel when all three percentages are between their respective lower and upper limits. The process of setting the color ratio range of the preset positioning sticker is limited to the sticker pattern itself. First, the red channel value, green channel value, and blue channel value of each standard color block in the printing draft of the positioning sticker are read. Then, the percentage of the three channels is calculated for each standard color block. The average percentage of each type of color block is used as the center value, and the center value is expanded by the same fixed deviation to form the corresponding channel percentage range. Next, adjacent identifier pixels are merged according to the condition that the difference between the horizontal coordinates is ≤1 and the difference between the vertical coordinates is ≤1. After merging, the average horizontal coordinate and the average vertical coordinate of each group of identifier pixels are calculated. The horizontal geometric center coordinate = the sum of the horizontal coordinates of all identifier pixels in the same group / the number of identifier pixels in the same group. The vertical geometric center coordinate = the sum of the vertical coordinates of all identifier pixels in the same group / the number of identifier pixels in the same group. The obtained geometric center coordinates are used as the coordinates of the candidate positioning point on the wall.

[0052] The parameter acquisition submodule establishes the point correspondence between the coordinates of the candidate positioning points on the wall and the coordinates of the standard feature points of the corresponding pre-set positioning sticker pattern. Through a random sampling consensus algorithm, it iteratively extracts a portion of the corresponding point pairs in the point correspondence, and obtains the perspective transformation parameters from the coordinates of the standard feature points of the positioning sticker pattern to the coordinates of the candidate positioning points on the wall in each iteration of the extracted portion of the corresponding point pairs, thus obtaining the perspective mapping parameter set.

[0053] The system reads a pre-defined coordinate table of standard feature points for the positioning sticker pattern and sorts the coordinates of candidate positioning points on the wall according to their horizontal and vertical positions and relative spacing. It then establishes a correspondence between the coordinates of each candidate positioning point and the coordinates of the standard feature points. During execution, it first selects a set of points from the candidate positioning points that matches the number of standard feature points, and then establishes a sequence number correspondence based on the arrangement order of the standard feature points. Subsequently, it repeatedly extracts a fixed number of corresponding point pairs from all corresponding point pairs. After each extraction, it lists the horizontal and vertical coordinates of the standard feature points in the extracted point pair and the horizontal and vertical coordinates of the candidate positioning points on the wall, and calculates a set of perspective transformation parameters based on the coordinate mapping relationship between point pairs with the same sequence number. When multiple points in the extraction result have overlapping horizontal and vertical coordinates, causing the projection relationship to be ununiquely determined, the extraction result is deleted. When the distribution of points in the extraction result can form an effective planar projection relationship, the extraction result is retained and the corresponding perspective transformation parameters are recorded. After all iterations are completed, a perspective mapping parameter set is formed.

[0054] The interior point retention submodule, based on the perspective mapping parameter group, maps the coordinates of the standard feature points of the positioning sticker pattern to the coordinate system of the wall image to obtain the theoretical projection coordinates of the wall. It calculates the two-dimensional plane straight-line distance between the coordinates of the corresponding candidate positioning points of the wall and the theoretical projection coordinates of the wall for each iteration, retains the coordinates of the candidate positioning points of the wall corresponding to a single iteration that are less than the preset distance judgment threshold and counts the number. It selects the set of candidate positioning point coordinates of the wall with the largest number as the wall reference positioning point.

[0055] In MATLAB, each set of perspective mapping parameters is applied sequentially to the coordinates of the standard feature points of the positioning sticker pattern, mapping these coordinates to the wall image coordinate system to obtain the theoretical projection coordinates of the wall corresponding to each standard feature point. After each mapping is completed, the coordinates of the corresponding candidate positioning points on the wall and the theoretical projection coordinates are read point by point. First, the horizontal and vertical coordinate deviations are calculated: Horizontal coordinate deviation = Horizontal coordinate of candidate positioning point - Horizontal coordinate of theoretical projection; Vertical coordinate deviation = Vertical coordinate of candidate positioning point - Vertical coordinate of theoretical projection. Then, the two-dimensional plane straight-line distance is calculated: [The two-dimensional plane straight-line distance is...]. Next, each distance value is compared with a preset distance judgment threshold. If the distance value is less than the preset distance judgment threshold, the point pair is retained; if the distance value is equal to the preset distance judgment threshold, the point pair is retained; if the distance value is greater than the preset distance judgment threshold, the point pair is deleted. After processing each set of perspective mapping parameters, the number of retained point pairs is counted. The coordinates of the candidate positioning point on the wall corresponding to the set with the largest number of retained point pairs are used as the wall reference positioning point. If there are two or more sets of retained point pairs with the same number, the total distance of each set of retained point pairs is calculated. The total distance is equal to the sum of the straight-line distances of all retained point pairs in the same set on the two-dimensional plane. The set with the smaller total distance is retained. If the total distance is still the same, the maximum single-point distance value in each set is compared. The set with the smaller maximum single-point distance value is retained.

[0056] The wall structure extraction module includes:

[0057] The grayscale conversion module obtains the brightness-weighted average of the red, green, and blue channel color values ​​of the wall to be paved, the corner area corresponding to the wall to be paved, and the door and window edge area from the wall pixel color data, and obtains the grayscale value of the wall image for each wall pixel.

[0058] For each pixel, the values ​​of the three channels are retrieved separately, and then calculated using the same set of brightness weights. The grayscale value of the wall image = red channel value × red brightness weight + green channel value × green brightness weight + blue channel value × blue brightness weight. The brightness weights are set based on the contribution relationship of visible light brightness, with the red, green, and blue brightness weights taking fixed values ​​and their sum equal to 1. The grayscale value obtained for each pixel is written into the grayscale table corresponding to its original coordinates. Grayscale values ​​between 0 and 51 are recorded as extremely dark, with 0 indicating no brightness response and 5 indicating no brightness response. 1 is approximately 1 / 5 of 255; values ​​between 52 and 102 are considered slightly dark, exceeding the extremely dark range but not yet reaching the middle range; values ​​between 103 and 153 are considered medium gray, falling within the middle of the overall range; values ​​between 154 and 204 are considered slightly bright, exceeding 154 and entering the upper range but not reaching the highest brightness; values ​​between 205 and 255 are considered extremely bright, exceeding 205 and approaching the upper limit of the range, with 255 representing the maximum brightness; after converting all pixels, the grayscale value sequence of the wall image used for subsequent edge selection is obtained.

[0059] The edge selection submodule calculates the horizontal and vertical brightness differences between the grayscale value of each wall pixel and the corresponding grayscale values ​​of its horizontal and vertical adjacent wall pixels. Based on the horizontal and vertical brightness differences, it determines the brightness variation range among multiple wall pixels and selects the coordinates of wall edge pixels whose brightness variation range exceeds the preset edge judgment threshold.

[0060] The wall pixels in the grayscale table are scanned point by point, and the grayscale values ​​of the wall image at the current pixel, the left adjacent pixel, the right adjacent pixel, the top adjacent pixel, and the bottom adjacent pixel are read sequentially. When the current pixel is inside the image, all four adjacent pixels (left, right, top, and bottom) are used in the calculation. When the current pixel is on the left edge of the image, the left adjacent pixel is missing, and only the right, top, and bottom adjacent pixels are used. When the current pixel is on the right edge of the image, the right adjacent pixel is missing, and only the left, top, and bottom adjacent pixels are used. When the current pixel is on the top edge of the image, the top adjacent pixel is missing. If the current pixel is missing, only the left, right, and bottom adjacent pixels are used; if the current pixel is located at the bottom edge of the image, the bottom adjacent pixels are missing, and only the left, right, and top adjacent pixels are used; then the horizontal and vertical brightness differences are calculated, where horizontal brightness difference = max{|current pixel gray value - left adjacent pixel gray value|, |current pixel gray value - right adjacent pixel gray value|}, and vertical brightness difference = max{|current pixel gray value - top adjacent pixel gray value|, |current pixel gray value - bottom adjacent pixel gray value|}; finally, the brightness change amplitude is calculated, where brightness change amplitude = Finally, the brightness change range is compared with the preset edge detection threshold. If the brightness change range is greater than the preset edge detection threshold, the current pixel coordinates are retained as the wall edge pixel coordinates; if the brightness change range is equal to the preset edge detection threshold, the current pixel coordinates are retained; if the brightness change range is less than the preset edge detection threshold, the current pixel coordinates are deleted.

[0061] The line conversion submodule uses the Hough transform algorithm to calculate the polar angle and polar radius values ​​of the wall edge pixel coordinates in the preset polar coordinate system, counts the cumulative number of parameter votes, selects the target polar angle and polar radius value combination that is higher than the preset line voting threshold, converts it into a wall line in the wall image coordinate system, and generates a wall physical structure line dataset.

[0062] The cumulative number of parameter votes is the weighted sum of the votes corresponding to the same polar angle value and polar radius value;

[0063] Converting the wall line to the wall image coordinate system involves determining the polar angle value in the combination of target polar angle and polar radius values ​​as the normal direction of the wall line, determining the polar radius value as the normal distance from the origin of the wall image coordinate system to the wall line, and converting the wall line to the wall image coordinate system based on the normal direction and normal distance of the wall line.

[0064] First, apply the original Hough linear relationship. Calculate the polar radius value of each wall edge pixel in the preset polar coordinate system, where... Indicates the first The polar radius values ​​corresponding to the polar angle values ​​are expressed in pixels. Indicates the first The horizontal coordinates of a pixel point on the wall edge, in pixels. Indicates the first The vertical coordinates of each pixel point on the wall edge, in pixels. Indicates the first A number of discrete polar angle values, in radians; polar angles are set from 0 to 179; based on the original Hough linear relationship, a weighted summation relationship is used for the cumulative number of parameter votes: ,in, Indicates the first The polar angle value and the first The cumulative number of parameter votes corresponding to each polar radius value This represents the total number of wall edge pixels that participated in the current polar angle and polar radius value statistics. This represents the pixel voting weight determined based on the brightness change amplitude corresponding to the pixel coordinates at the wall edge, with a value ranging from 0 to 1. Indicates the first The coordinates of the pixel point on the wall edge are at the... The polar radius value is calculated based on the polar angle value, and the unit is pixels. Indicates the first A discrete polar radius value, in pixels. This represents the polar radius tolerance width set based on a preset polar coordinate system, in pixels; where, , Indicates the first The brightness variation range corresponding to each pixel on the wall edge is derived from the previously calculated brightness variation range. This represents the maximum value among all the brightness variations of pixels at the edge of the wall. At that time, the cumulative number of parameter votes for the current polar angle value and polar radius value is directly recorded as 0; Take 1 to 3 pixels; taking the coordinates of a single pixel on the edge of a wall as an example, the coordinates are known. Pixels Pixels Convert to radians and substitute. ,get Pixels; then take Pixels , ,but ,Pick Pixels; Substitute The corresponding terms for the current pixel, excluding those obtained through summation, are obtained. After all wall edge pixels have undergone the same calculation, each combination of polar angle and polar radius values ​​is accumulated separately. This involves fixing one polar angle value and one polar radius value, and only adding the voting results calculated for all wall edge pixels under that polar angle value and corresponding to that polar radius value to obtain the cumulative number of parameter votes for that combination. Then, the polar angle or polar radius value is changed to form the next combination, and the cumulative number of parameter votes for each combination is calculated. After completing the pairing and statistics of all polar angle values ​​and all polar radius values, each combination of "polar angle value + polar radius value" corresponds to an independent cumulative number of parameter votes. Finally, the cumulative number of parameter votes corresponding to each combination of polar angle and polar radius values ​​is compared with the preset straight line vote. Thresholds are compared group by group. When the cumulative number of parameter votes is greater than the preset line voting threshold, the group of polar angle and polar radius value combinations is retained as the target polar angle and polar radius value combinations. When the cumulative number of parameter votes is equal to the preset line voting threshold, the group of polar angle and polar radius value combinations is retained as the target polar angle and polar radius value combinations. When the cumulative number of parameter votes is less than the preset line voting threshold, the group of polar angle and polar radius value combinations is deleted. Then, the retained target polar angle and polar radius value combinations are converted into wall lines in the wall image coordinate system, where the polar angle value is used as the normal direction of the wall line and the polar radius value is used as the normal distance from the image coordinate origin to the wall line. Then, the coordinates of all wall edge pixels corresponding to the polar angle and polar radius value combinations are written into the same line record to form a wall physical structure line dataset.

[0065] Compared to the traditional Hough transform, which calculates weighted sums based solely on whether an edge pixel falls within the corresponding polar angle-radius parameter, this formula introduces pixel voting weights determined by the magnitude of brightness variation during parameter voting accumulation. This allows wall edge pixels with more pronounced brightness changes and more prominent edge features to contribute higher voting values, while the voting contributions of weak edges and noisy edges are correspondingly reduced. Furthermore, instead of a hard-score method of "one vote for complete overlap and zero votes for non-overlap," it uses an exponential decay term to continuously attenuate and weight the deviation between the calculated and discrete polar radius values ​​of pixels. This allows pixels close to the target polar radius to participate in accumulation based on their proximity, thereby enhancing the parameter statistics' tolerance to polar radius quantization errors, edge discrete fluctuations, and local noise interference.

[0066] The baseline determination module includes:

[0067] The direction fitting submodule combines the arrangement direction of the wall reference positioning points and the standard feature points of the positioning sticker pattern to fit the horizontal direction vector and the vertical direction vector of the wall, obtain the direction vector of the wall line in the wall physical structure straight line data, and calculate the cosine value of the angle between the straight line direction vector and the vertical direction vector of the wall.

[0068] After generating the linear dataset of the wall's physical structure, the arrangement relationship between the wall's reference positioning points and the standard feature points of the positioning sticker pattern is read together. The vertical and horizontal vectors of the wall are extracted according to the vertical strip splicing method. During execution, first, two points in the same column of the standard feature points are selected, and then two corresponding points are selected from the wall's reference positioning points. The horizontal and vertical coordinate differences between the two pairs of points are calculated, and the coordinate difference formed by the wall's reference positioning point pairs is determined as the wall's vertical vector. Then, two points in the same row of the standard feature points are selected, and two corresponding points are selected from the wall's reference positioning points. The coordinate difference formed by the wall's reference positioning point pairs is determined as the wall's horizontal vector. Then, the coordinates of the two endpoints of each straight line on the wall are retrieved one by one from the wall physical structure straight line dataset. The difference between the horizontal and vertical coordinates of the endpoints is calculated to obtain the direction vector of the straight line. Next, the cosine of the angle between the direction vector of the straight line and the vertical direction vector of the wall is calculated. The cosine of the angle is calculated as follows: (horizontal component of the direction vector of the straight line × horizontal component of the vertical direction vector of the wall + vertical component of the direction vector of the straight line × vertical component of the vertical direction vector of the wall) / (length of the direction vector of the straight line × length of the vertical direction vector of the wall). When the cosine of the angle is in the range of 0.95 to 1, it is considered to be approximately parallel. When the cosine of the angle is in the range of 0.8 to 0.95, it is considered to have a small directional deviation. When the cosine of the angle is < 0.8, it is considered to have a large directional deviation.

[0069] The baseline screening module selects wall lines with a cosine value of the included angle greater than a preset parallelism threshold as vertical reference lines for wall sound insulation materials. At the same time, it selects wall lines from the wall physical structure line data set that are parallel to the horizontal vector of the wall and located at the lower boundary of the corner area corresponding to the wall to be laid as horizontal lines at the bottom boundary of the wall.

[0070] At key screening locations, MATLAB is used to retrieve the cosine value of the included angle of each straight line in the wall physical structure straight line dataset and compare it with a preset parallelism threshold. If the cosine value is greater than the preset parallelism threshold, the wall line is retained and added to the vertical candidate line set; if the cosine value is equal to the preset parallelism threshold, the wall line is retained and added to the vertical candidate line set; if the cosine value is less than the preset parallelism threshold, the wall line is deleted. The setting process of the preset parallelism threshold is limited to the angular deviation range between the wall line direction and the vertical laying direction. First, an upper limit of the allowable angular deviation is given, and then the cosine value of the included angle corresponding to the upper limit of the allowable angular deviation is used as the preset parallelism threshold. Subsequently, the lateral distance from each straight line to the corner of the wall on the starting side is calculated in the vertical candidate line set. The lateral distance is the maximum. The smallest straight line is retained as the vertical reference line for the wall sound insulation material. Then, the wall straight lines parallel to the horizontal vector of the wall are further filtered from the wall physical structure straight line dataset. The parallelism determination adopts the same cosine value comparison method as above. Straight lines that reach the horizontal parallelism threshold are retained, and those below the horizontal parallelism threshold are deleted. The vertical position of the retained horizontal straight line is read and compared with the lower boundary position of the corner area corresponding to the wall to be laid. The vertical coordinate deviation = vertical position of the candidate horizontal straight line - vertical position of the lower boundary of the corner area. When the vertical coordinate deviation is within the preset allowable range, it is retained as the horizontal straight line of the bottom boundary of the wall. When the vertical coordinate deviation is lower than the lower limit of the preset allowable range, it is deleted. When the vertical coordinate deviation is higher than the upper limit of the preset allowable range, it is deleted.

[0071] The coordinate translation submodule sets the coordinates of the intersection point of the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall as the origin coordinates of the wall sound insulation material. At the same time, starting from the origin coordinates of the wall sound insulation material, it translates at equal distances along the horizontal vector direction of the wall according to the preset standard width value of the wall sound insulation material to determine the baseline coordinates of the wall sound insulation material.

[0072] The geometric relationship between the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall is read. The coordinates of the intersection point of the two lines are obtained, and the horizontal and vertical coordinates of the intersection point are written into the starting origin table as the starting origin coordinates of the wall sound insulation material. Then, the preset standard width value of the wall sound insulation material is read, and the known actual distance between the standard feature points of the positioning sticker and their corresponding pixel distance in the wall image coordinate system are read. The pixel size ratio is calculated as follows: pixel size ratio = actual distance between standard feature points / distance between corresponding pixels of standard feature points. Then, the standard width value of the wall sound insulation material is converted into the translation pixel distance in the image: translation pixel distance = standard width value of wall sound insulation material / pixel size ratio. Subsequently, the following steps are taken: Starting from the coordinates of the origin point where the wall sound insulation material is laid, the material is successively translated at equal distances along the horizontal direction of the wall. The first laying baseline passes through the origin point. The second laying baseline is translated by 1 pixel relative to the first laying baseline along the horizontal direction of the wall. The third laying baseline is translated by another 1 pixel relative to the second laying baseline, and so on. Each time a new laying baseline is obtained, it is determined whether the laying baseline is still within the range of the wall to be laid. If the laying baseline is completely within the range of the wall to be laid, it is retained. If the laying baseline coincides with the boundary of the wall, it is retained. If the laying baseline exceeds the range of the wall to be laid, the laying baseline and its subsequent translation results are deleted. Thus, the coordinates of the wall sound insulation material laying baseline are obtained.

[0073] The location map generation module includes:

[0074] The node determination submodule refers to the wall physical structure straight line dataset, filters the wall straight lines located in the door and window edge line area, and determines the door and window boundary straight lines according to the extension direction and position continuity of the wall straight lines in the door and window edge line area. Based on the coordinates of the door and window boundary straight lines and the baseline coordinates of the wall sound insulation material laying, the coordinates of the two-dimensional plane intersection point are determined, which are used as the coordinates of the wall sound insulation material cutting node.

[0075] After the baseline coordinates are determined, wall lines located within the door and window edge areas are retrieved from the wall physical structure straight line dataset, and the extension direction and positional continuity of each line are compared one by one. During execution, the direction difference and endpoint spacing of any two candidate lines are calculated first. The direction difference is the absolute value of the difference between the direction angles of the two candidate lines. The difference in the squares of the endpoint lateral coordinates is the square of (the lateral coordinate of the first candidate line endpoint - the lateral coordinate of the second candidate line endpoint). The difference in the squares of the endpoint longitudinal coordinates is the square of (the longitudinal coordinate of the first candidate line endpoint - the longitudinal coordinate of the second candidate line endpoint). The endpoint spacing is... When the directional difference is within a preset continuous directional interval and the endpoint spacing is within a preset continuous distance interval, the two candidate lines are grouped into the same door and window boundary group; when the directional difference exceeds the preset continuous directional interval, they are grouped into different door and window boundary groups; when the endpoint spacing exceeds the preset continuous distance interval, they are grouped into different door and window boundary groups; after grouping, the candidate lines in each group are arranged in a continuous positional order and connected to form door and window boundary lines; then, each door and window boundary line is intersected with each wall sound insulation material laying baseline one by one. Before intersecting, the directional relationship between the two lines is determined. If they are parallel, no intersection point is generated; if they overlap, the start and end points of the overlapping segment are taken as cutting nodes; if they are not parallel, a unique intersection point is calculated; for each intersection point obtained, it is further determined whether the intersection point is simultaneously within the effective length range of the door and window boundary line and within the range of the corresponding laying baseline line segment. If both conditions are met, it is retained as the wall sound insulation material cutting node coordinates; if only one condition is met, it is deleted; if neither condition is met, it is deleted.

[0076] The scaling conversion submodule calls the screen display size scaling factor to adjust the positional relationship between the starting point coordinates of the wall sound insulation material, the laying baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, and calculates the display scaling coordinate group.

[0077] The scaling factor for screen display size is determined based on the corresponding ratio between the pixel size in the coordinate system of the wall image and the display size in the coordinate system of the display interface;

[0078] The system retrieves the width and height pixel values ​​of the display interface, as well as the image width and height pixel values ​​in the wall image coordinate system, and calculates the horizontal and vertical scaling ratios respectively. The horizontal scaling ratio is calculated as: horizontal scaling ratio = display interface width pixel value / wall image width pixel value; the vertical scaling ratio is calculated as: display interface height pixel value / wall image height pixel value. Then, it reads the coordinates of the wall sound insulation material's starting point, baseline, and cutting node, and performs scaling conversion on each coordinate. The display horizontal coordinate is calculated as: original horizontal coordinate × scaling ratio + horizontal offset; the display vertical coordinate is calculated as: original vertical coordinate × scaling ratio + vertical offset. When the horizontal scaling ratio equals the vertical scaling ratio, the same ratio is used for scaling. The calculation is as follows: horizontal offset = 0, vertical offset = 0; when the horizontal scaling ratio > the vertical scaling ratio, the vertical scaling ratio is used for unified conversion: horizontal offset = (display interface width in pixels - wall image width in pixels × vertical scaling ratio) / 2, vertical offset = 0; when the horizontal scaling ratio < the vertical scaling ratio, the horizontal scaling ratio is used for unified conversion: horizontal offset = 0, vertical offset = (display interface height in pixels - wall image height in pixels × horizontal scaling ratio) / 2; after the conversion, it is further determined whether the scaling coordinates are within the display interface boundary. If the scaling coordinates are within the display interface boundary, they are retained; if the scaling coordinates are on the display interface boundary line, they are retained; if the scaling coordinates exceed the display interface boundary, they are deleted. This results in the display scaling coordinate group.

[0079] The image construction submodule, based on the display scaling coordinate group, stitches and reassembles the display scaling coordinates corresponding to the starting point coordinates of the wall sound insulation material, the baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, to generate a positioning map of the wall sound insulation material based on the wall image;

[0080] Finally, at the key display construction location, MATLAB is called to create a 2D canvas with the same size as the display interface, and then the coordinates of the starting point, baseline, and cutting nodes of the wall sound insulation material are sequentially called from the display scaling coordinate group. First, the starting point coordinates are written into the canvas, then vertical laying lines are generated by connecting the starting and ending points of each laying baseline. Subsequently, the cutting nodes falling on the same vertical laying line are sorted by their vertical coordinates from smallest to largest and written into the laying line point by point. When a laying line does not intersect with the boundary of a door or window, the entire laying line is displayed; when a certain laying line does not intersect with the boundary of a door or window, the entire laying line is displayed. When a laying line has one or more cutting nodes, the laying line is divided into multiple segments according to the positional relationship between adjacent cutting nodes. When the boundary of a door or window coincides with the laying line to form an overlapping segment, the start and end points of the overlapping segment are written into the canvas as the cutting positions. When the coordinates of a cutting node coincide with the coordinates of the endpoint of the laying line, the cutting node is written into the endpoint position of the laying line. When the coordinates of a cutting node are not within the range of the line segment, it is deleted. After all the laying lines and cutting nodes are written, a wall sound insulation material laying positioning map based on the wall image is formed. The aforementioned starting point, laying baseline for each section, and corresponding cutting nodes are kept in correspondence in the same canvas.

[0081] When the system is executed, it first acquires the wall image, then identifies the positioning sticker and wall structure, then determines the starting point, laying baseline, and door and window cutting nodes, and finally displays the result as a positioning diagram that can be directly referenced for construction. Since wall sound insulation materials can be laid in whole panels, horizontally spliced, or vertically spliced, and this solution generates multiple vertical laying baselines by successively shifting according to the standard width, it corresponds to the construction of vertical strip panels spliced ​​one by one, and does not correspond to the construction method of covering the wall with a single whole panel at once. In the actual construction of wall sound insulation materials, the construction personnel can directly determine where to start laying, where each piece of material will be laid, and where to cut at door and window locations based on the wall sound insulation material laying positioning diagram, thereby reducing the need for repeated manual measurement and layout.

[0082] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A wall sound insulation material laying auxiliary positioning system based on wall images, characterized in that the system... include: The wall image acquisition module acquires the color data of wall pixels covering the wall to be covered, the corner area of ​​the wall to be covered, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall. The positioning sticker recognition module determines the coordinates of candidate positioning points on the wall based on the color data of the wall pixels and the pattern area of ​​the positioning sticker, and then filters the wall reference positioning points based on the coordinates of the candidate positioning points. The wall structure extraction module determines the grayscale value of the wall image and the coordinates of the wall edge pixels by referring to the wall pixel color data, and summarizes the straight line dataset of the wall physical structure based on the wall edge pixel coordinates. The baseline determination module determines the vertical reference line of the wall sound insulation material based on the wall reference positioning point, and selects the horizontal straight line of the bottom boundary of the wall based on the straight line dataset of the wall physical structure. Based on the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall, the coordinates of the starting point of the wall sound insulation material and the coordinates of the wall sound insulation material laying baseline are set. The positioning map generation module determines the coordinates of the wall sound insulation material cutting nodes based on the straight line dataset of the wall physical structure and the baseline coordinates of the wall sound insulation material laying. Based on the coordinates of the starting point coordinates of the wall sound insulation material, the baseline coordinate data of the wall sound insulation material laying, and the coordinates of the wall sound insulation material cutting nodes, it constructs a positioning map of the wall sound insulation material laying.

2. The wall sound insulation material laying auxiliary positioning system based on wall image as described in claim 1, characterized in that, The wall image acquisition module includes: The color extraction submodule acquires a colored wall image covering the wall to be paved, the corner area corresponding to the wall to be paved, the door and window edge area, and the positioning sticker pattern pre-pasted on the wall. It scans the wall pixels in the colored wall image line by line and extracts the red channel color value, green channel color value, and blue channel color value corresponding to each wall pixel to obtain a three-channel color value group of the wall pixel. The region alignment submodule establishes a correspondence between the pixel coordinates and three-channel color values ​​of the wall surface, the corner area corresponding to the wall surface, the door and window edge area, and the positioning sticker pattern area based on the three-channel color value group of the wall surface pixels, and obtains the region color correspondence set. The color combination submodule extracts the red, green, and blue channel color values ​​corresponding to the pixel coordinates of each region in the region color correspondence set. Based on the horizontal and vertical arrangement of the pixels, it establishes a wall image coordinate system with the top left pixel as the origin, the horizontal direction as the horizontal coordinate direction, and the vertical direction as the vertical coordinate direction, and generates wall pixel color data.

3. The wall sound insulation material laying auxiliary positioning system based on wall image as described in claim 1, characterized in that, The location sticker recognition module includes: The sticker screening module selects the coordinates of the marker pixels whose color ratios match the preset color ratio range of the marker pixels based on the red channel color values, green channel color values, and blue channel color values ​​of the marker pixel pattern area in the wall pixel color data. It merges multiple sets of adjacent marker pixel coordinates and calculates the geometric center coordinates of each set of marker pixel coordinates as the coordinates of the candidate positioning points on the wall. The parameter acquisition submodule establishes the point correspondence between the coordinates of the candidate positioning points on the wall and the coordinates of the standard feature points of the corresponding pre-set positioning sticker pattern. Through a random sampling consensus algorithm, it iteratively extracts a portion of the corresponding point pairs in the point correspondence, and obtains the perspective transformation parameters from the coordinates of the standard feature points of the positioning sticker pattern to the coordinates of the candidate positioning points on the wall in each iteration of the extracted portion of the corresponding point pairs, thus obtaining the perspective mapping parameter set. The interior point retention submodule, based on the perspective mapping parameter group, maps the coordinates of the standard feature points of the positioning sticker pattern to the wall image coordinate system to obtain the theoretical projection coordinates of the wall. It calculates the two-dimensional plane straight-line distance between the coordinates of the corresponding candidate positioning points on the wall and the theoretical projection coordinates of the wall for each iteration, retains the coordinates of the candidate positioning points on the wall corresponding to a single iteration that are less than a preset distance judgment threshold and counts the number. It selects the set of candidate positioning point coordinates on the wall with the largest number as the wall reference positioning point.

4. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 3, characterized in that, The wall structure extraction module includes: The grayscale conversion module obtains the brightness-weighted average of the red, green, and blue channel color values ​​of the wall to be paved, the corner area corresponding to the wall to be paved, and the door and window edge area in the wall pixel color data, and obtains the grayscale value of the wall image of each wall pixel. The edge selection submodule calculates the horizontal and vertical brightness differences between the grayscale value of each wall pixel and the corresponding grayscale values ​​of its horizontal and vertical adjacent wall pixels. Based on the horizontal and vertical brightness differences, it determines the brightness variation range among multiple wall pixels and selects the coordinates of wall edge pixels whose brightness variation range exceeds a preset edge determination threshold. The straight line conversion submodule calculates the polar angle and polar radius values ​​of the wall edge pixel coordinates in the preset polar coordinate system using the Hough transform algorithm, counts the cumulative number of parameter votes, selects the target polar angle and polar radius value combination that is higher than the preset straight line voting threshold, converts it into a straight line of the wall in the wall image coordinate system, and generates a straight line dataset of the wall physical structure.

5. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 4, characterized in that, The laying baseline determination module includes: The direction fitting submodule combines the arrangement direction of the wall reference positioning points and the standard feature points of the positioning sticker pattern to fit the horizontal direction vector and the vertical direction vector of the wall, obtain the direction vector of the wall line in the wall physical structure straight line data, and calculate the cosine value of the angle between the straight line direction vector and the vertical direction vector of the wall. The baseline screening module selects wall straight lines whose included angle cosine value is greater than a preset parallelism threshold as vertical reference lines for wall sound insulation materials. At the same time, it selects wall straight lines from the wall physical structure straight line dataset that are parallel to the horizontal vector of the wall and located at the lower boundary of the corner area corresponding to the wall to be laid as horizontal straight lines at the bottom boundary of the wall. The coordinate translation submodule sets the coordinates of the intersection point of the vertical reference line of the wall sound insulation material and the horizontal straight line of the bottom boundary of the wall as the origin coordinates of the wall sound insulation material. At the same time, starting from the origin coordinates of the wall sound insulation material, it translates at equal distances along the horizontal vector direction of the wall according to the preset standard width value of the wall sound insulation material to determine the baseline coordinates of the wall sound insulation material.

6. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 5, characterized in that, The location map generation module includes: The node determination submodule, referring to the wall physical structure straight line dataset, filters the wall straight lines located in the door and window edge line area, and determines the door and window boundary straight lines according to the extension direction and position continuity of the wall straight lines in the door and window edge line area. Based on the door and window boundary straight lines and the baseline coordinates of the wall sound insulation material laying, it determines the coordinates of the two-dimensional plane intersection point, which is used as the coordinates of the wall sound insulation material cutting node. The scaling conversion submodule calls the screen display size scaling factor to adjust the positional relationship between the starting point coordinates of the wall sound insulation material, the laying baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, and calculates the display scaling coordinate group. The image construction submodule, based on the display scaling coordinate group, splices and reassembles the display scaling coordinates corresponding to the starting point coordinates of the wall sound insulation material, the baseline coordinates of the wall sound insulation material, and the cutting node coordinates of the wall sound insulation material, to generate a positioning map of the wall sound insulation material based on the wall image.

7. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 3, characterized in that, The color ratio range of the positioning sticker includes the proportion range of the red channel, the proportion range of the green channel, and the proportion range of the blue channel.

8. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 4, characterized in that, The cumulative number of parameter votes is the weighted sum of the voting results corresponding to the same polar angle value and polar radius value.

9. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 4, characterized in that, The conversion to a straight line in the wall image coordinate system includes determining the polar angle value in the target polar angle and polar radius combination as the normal direction of the straight line, determining the polar radius value as the normal distance from the origin of the wall image coordinate system to the straight line, and converting the straight line in the wall image coordinate system based on the normal direction and normal distance of the straight line.

10. The wall sound insulation material laying auxiliary positioning system based on wall image according to claim 6, characterized in that, The screen display size scaling factor is determined based on the corresponding ratio between the pixel size of the wall image coordinate system and the display size of the display interface coordinate system.

Images