House type structure diagram generation method

Abstract

The embodiment of the present disclosure discloses a method for generating a house layout structure diagram, which comprises the following steps: obtaining first point cloud data information of a target house and trajectory information of a collecting device of the first point cloud data, wherein the first point cloud data information is original point cloud data information of the target house; determining a house layout structure diagram of the target house according to the first point cloud data information and the trajectory information; and outputting the house layout structure diagram of the target house. Thus, the structure data of the target house is obtained through the point cloud data, the house layout structure diagram is automatically drawn, and the accuracy of house measurement is improved.

Description

Methods for generating house floor plan diagrams Technical Field

[0001] This disclosure relates to three-dimensional reconstruction technology, point cloud processing technology, and in particular a method for generating a house floor plan structure diagram. Background Technology

[0002] In traditional house measurement, surveyors typically use measuring tapes or laser rangefinders, combined with their understanding of the entire house's space, to first sketch the house, then record the measured dimensions, and finally draw the entire house in CAD (Computer-Aided Drafting) software, creating a floor plan. Traditional house measurement has several problems. For example, it requires a certain level of expertise from the surveyor, demanding precise measurements without overlooking any details to create a complete CAD drawing. Furthermore, if the measurement records contain errors, the surveyor must return to the site to remeasure. In cases where remeasurement is not possible, corrections must be made based on experience. Therefore, the accuracy of traditional house measurement is difficult to guarantee. Summary of the Invention

[0003] To address the technical problems in related technologies, this disclosure provides a method for generating house floor plan diagrams.

[0004] According to a first aspect of the present disclosure, a method for generating a house floor plan is provided, the method comprising:

[0005] Acquire the first point cloud data information of the target house and the trajectory information of the device that collects the first point cloud data;

[0006] Based on the first point cloud data information and trajectory information, determine the house layout of the target house;

[0007] Output the floor plan of the target house.

[0008] As an optional embodiment of this disclosure, determining the house layout diagram of the target house based on the first point cloud data information and trajectory information includes:

[0009] The first point cloud data information is denoised to obtain the second point cloud data information;

[0010] The spatial structure information of the target house is determined based on the second point cloud data information;

[0011] Based on the spatial structure information of the target house, the internal and external contour information of the target house is determined, and the first house layout diagram of the target house is obtained;

[0012] Based on the first floor plan of the target house, the window structure of the target house is determined to obtain the second floor plan, which is the final floor plan of the target house.

[0013] As an optional embodiment of this disclosure, the step of denoising the first point cloud data information to obtain the second point cloud data information includes:

[0014] The first noisy point cloud in the first point cloud data information is removed by using a preset point cloud dynamic object removal algorithm to obtain the second point cloud data information. The first noisy point cloud is dynamic object noise information.

[0015] As an optional embodiment of this disclosure, the step of denoising the first point cloud data information to obtain the second point cloud data information further includes:

[0016] The coordinate range of the mirror plane is determined from the first point cloud data information;

[0017] The point cloud data information in the first point cloud data information that falls within the coordinate range of the mirror plane is determined as the second noise point cloud;

[0018] The second noisy point cloud is deleted from the first point cloud data information to obtain the second point cloud data information.

[0019] As an optional embodiment of this disclosure, the step of denoising the first point cloud data information to obtain the second point cloud data information further includes:

[0020] The first point cloud data information is projected onto a first preset plane and then onto a preset grid to obtain a first occupied grid map, wherein the first preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud;

[0021] Project the trajectory information of the first point cloud data acquisition device onto a first preset plane to obtain the two-dimensional trajectory information of the acquisition device on the first preset plane;

[0022] All connected components formed by the first point cloud data information are determined based on the first occupancy raster map;

[0023] Based on the two-dimensional trajectory information of the acquisition device in the first preset plane, the noisy connected component is determined from all connected components;

[0024] The point cloud data information in the first point cloud data information that is in the noise connected domain is determined as the third noise point cloud;

[0025] The third noisy point cloud is deleted from the first point cloud data information to obtain the second point cloud data information.

[0026] As an optional embodiment of this disclosure, the step of denoising the first point cloud data information to obtain the second point cloud data information further includes:

[0027] Before projecting the first point cloud data information onto the first preset plane into the first preset grid to obtain the first occupied grid map, the normal of the first point cloud data information is adjusted according to the first average normal distribution of the first point cloud data information to obtain the second average normal distribution of the first point cloud data information, wherein the second average normal distribution of the first point cloud data information is perpendicular to the first preset plane.

[0028] As an optional embodiment of this disclosure, the step of denoising the first point cloud data information to obtain the second point cloud data information further includes:

[0029] Based on the first point cloud data information, determine two planes of any wall of the target house;

[0030] The point cloud data information in the first point cloud data information that is between two planes on any wall of the target house is determined as the fourth noise point cloud;

[0031] The fourth noisy point cloud is deleted from the first point cloud data information to obtain the second point cloud data information.

[0032] As an optional embodiment of this disclosure, determining the spatial structure information of the target house based on the second point cloud data information includes:

[0033] The wall plane of the target house is determined based on the second point cloud data information;

[0034] The first outer rectangular outline is determined based on the bounding box of the second point cloud data information;

[0035] The spatial structure information of the target house is determined using the wall plane and the first outer rectangular outline.

[0036] As an optional embodiment of this disclosure, determining the wall plane of the target house based on the second point cloud data information includes:

[0037] Using the region growing method and the region average normal, all spatial planes of the target house are determined based on the second point cloud data information;

[0038] The vertical plane is determined from all spatial planes of the target house using a preset plane equation;

[0039] The second point cloud data information is projected onto the second preset plane and then onto the second preset grid to obtain the second occupied grid map, wherein the second preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud;

[0040] The vertical plane located inside the second occupied grid is deleted to obtain the vertical plane located at the edge of the second occupied grid.

[0041] The vertical plane located at the edge of the second occupancy grid is defined as the wall plane of the target house.

[0042] As an optional embodiment of this disclosure, determining the spatial structure information of the target house using the wall plane and the first outer rectangular outline includes:

[0043] The outline size of the first outer rectangular outline is enlarged and adjusted to obtain the second outer rectangular outline;

[0044] The point cloud data of the wall plane is projected onto the second preset plane to obtain the wall surface line segment of the wall plane;

[0045] The spatial structure information of the target house is determined by using the wall line segments of the wall plane and the second outer rectangular outline.

[0046] As an optional embodiment of this disclosure, determining the spatial structure information of the target house using the wall surface segments of the wall plane and the second outer rectangular outline includes:

[0047] Based on the position coordinates of the wall line segments on the wall plane, the wall line segments are added to the corresponding coordinate positions of the second outer rectangular outline to obtain the initial spatial structure of the target house.

[0048] Each wall segment in the initial spatial structure of the target house is supplemented with additional wall segments to obtain the supplemented wall segments.

[0049] The final spatial structure formed by the supplemented wall segments and the second outer rectangular outline is determined as the spatial structure information of the target house.

[0050] As an optional embodiment of this disclosure, the step of determining the inner and outer contour information of the target house based on the spatial structure information of the target house to obtain a first house layout diagram of the target house includes:

[0051] Based on the spatial structure information of the target house, a corresponding image to be segmented is determined, wherein the vertices of the image to be segmented are the polygons formed by the spatial structure information of the target house, and the edges of the image to be segmented include the common edges of adjacent polygons formed by the spatial structure information of the target house and the edges formed by connecting the vertices of the image to be segmented with the source and sink points of the graph cut respectively.

[0052] The weight of each edge in the graph to be segmented is determined based on the vertices, edges, source and sink vertices of the graph cut.

[0053] The minimum cut of the graph to be segmented is determined by calculating the weight of each edge in the graph based on a preset graph cut algorithm, thereby obtaining the internal and external boundaries of the spatial structure of the target house.

[0054] By connecting the internal and external boundaries of the spatial structure of the target house, a first house layout diagram of the target house is obtained.

[0055] As an optional embodiment of this disclosure, the step of connecting the internal and external boundaries of the spatial structure of the target house to obtain a first floor plan of the target house includes:

[0056] The polygonal outline of the target house is obtained by concatenating the edges of the internal and external boundaries of the spatial structure of the target house.

[0057] The first target merging vertex is determined from any two vertices of the polygonal contour that meet the first preset merging condition, or the second target merging vertex is determined from multiple vertices of the polygonal contour that meet the second preset merging condition.

[0058] Merging the first target vertices or the second target vertices is performed to obtain a merged polygonal outline, which is the first house layout diagram of the target house.

[0059] As an optional embodiment of this disclosure, the step of determining the window structure of the target house based on the first floor plan of the target house to obtain a second floor plan, wherein the second floor plan is the final floor plan of the target house, includes:

[0060] Obtain the point cloud data information corresponding to the edges of each contour in the first house floor plan;

[0061] The point cloud data information corresponding to the edge of each contour is projected onto the third preset plane and then onto the third preset grid to obtain the third occupied grid map. The third preset plane is the plane formed by the vertical axis and the vertical axis in the three-dimensional coordinate system of the point cloud.

[0062] For any edge of a contour, determine whether it includes a hole region in the corresponding third occupancy grid.

[0063] If the hollow area is determined to be included, then the rectangular bounding box of the hollow area is determined, and the size and coordinate position of the hollow area are obtained. The hollow area is the edge window structure of the corresponding outline in the first house floor plan.

[0064] According to the size and coordinate position of the hollow area, draw the hollow area on the edge of the corresponding outline in the first house floor plan to obtain the second house floor plan.

[0065] According to a second aspect of the present disclosure, an apparatus for generating a house floor plan is provided, the apparatus comprising:

[0066] The data acquisition module is used to acquire the first point cloud data information of the target house and the trajectory information of the acquisition device for the first point cloud data;

[0067] The house layout diagram determination module is used to determine the house layout diagram of the target house based on the first point cloud data information and trajectory information;

[0068] The house floor plan output module is used to output the house floor plan of the target house.

[0069] As an optional embodiment of this disclosure, the house floor plan determination module includes:

[0070] A noise reduction processing unit is used to perform noise reduction processing on the first point cloud data information to obtain the second point cloud data information.

[0071] The first determining unit is used to determine the spatial structure information of the target house based on the second point cloud data information;

[0072] The second determining unit is used to determine the inner and outer contour information of the target house based on the spatial structure information of the target house, so as to obtain the first house layout diagram of the target house;

[0073] The third determining unit is used to determine the window structure of the target house based on the first house layout diagram of the target house, and obtain a second house layout diagram, which is the final house layout diagram of the target house.

[0074] As an optional embodiment of this disclosure, the noise reduction processing unit includes:

[0075] The first noise removal subunit is used to remove the first noise point cloud from the first point cloud data information using a preset point cloud dynamic object removal algorithm to obtain the second point cloud data information, wherein the first noise point cloud is dynamic object noise information.

[0076] As an optional embodiment of this disclosure, the noise reduction processing unit further includes:

[0077] The mirror plane determination subunit is used to determine the coordinate range of the mirror plane from the first point cloud data information;

[0078] The first noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information that is within the coordinate range of the mirror plane as the second noise point cloud.

[0079] The second noise removal subunit deletes the second noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0080] As an optional embodiment of this disclosure, the noise reduction processing unit further includes:

[0081] The first projection subunit projects the first point cloud data information onto a first preset plane onto a preset grid to obtain a first occupied grid map, wherein the first preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud;

[0082] The second projection subunit is used to project the trajectory information of the acquisition device of the first point cloud data onto the first preset plane to obtain the two-dimensional trajectory information of the acquisition device on the first preset plane.

[0083] The connected component determination subunit is used to determine all connected components formed by the first point cloud data information based on the first occupancy raster map.

[0084] The noise connectivity determination subunit is used to determine the noise connectivity from all the connectivity based on the two-dimensional trajectory information of the acquisition device on the first preset plane.

[0085] The second noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information and in the noise connected domain as the third noise point cloud.

[0086] The third noise removal subunit is used to delete the third noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0087] As an optional embodiment of this disclosure, the noise reduction processing unit further includes:

[0088] The second projection subunit is used to adjust the normal of the first point cloud data information according to the first average normal distribution of the first point cloud data information before projecting the first point cloud data information onto the first preset plane onto the first preset grid to obtain the first occupied grid map, thereby obtaining the second average normal distribution of the first point cloud data information, wherein the second average normal distribution of the first point cloud data information is perpendicular to the first preset plane.

[0089] As an optional embodiment of this disclosure, the noise reduction processing unit further includes:

[0090] The first wall surface determination subunit is used to determine two planes of any wall of the target house based on the first point cloud data information.

[0091] The third noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information that is between two planes on any wall of the target house as the fourth noise point cloud;

[0092] The fourth noise removal subunit is used to delete the fourth noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0093] As an optional embodiment of this disclosure, the first determining unit includes:

[0094] The second wall surface determination subunit is used to determine the wall plane of the target house based on the second point cloud data information;

[0095] The rectangular outline determination subunit is used to determine the first outer rectangular outline based on the bounding box of the second point cloud data information.

[0096] The spatial structure determination subunit is used to determine the spatial structure information of the target house using the wall plane and the first outer rectangular outline.

[0097] As an optional embodiment of this disclosure, the second wall surface determining subunit is further configured to:

[0098] Using the region growing method and the region average normal, all spatial planes of the target house are determined based on the second point cloud data information;

[0099] The vertical plane is determined from all spatial planes of the target house using a preset plane equation;

[0100] The second point cloud data information is projected onto the second preset plane and then onto the second preset grid to obtain the second occupied grid map, wherein the second preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud;

[0101] The vertical plane located inside the second occupied grid is deleted to obtain the vertical plane located at the edge of the second occupied grid.

[0102] The vertical plane located at the edge of the second occupancy grid is defined as the wall plane of the target house.

[0103] As an optional embodiment of this disclosure, the spatial configuration determining sub-unit is further used for:

[0104] The outline size of the first outer rectangular outline is enlarged and adjusted to obtain the second outer rectangular outline;

[0105] The point cloud data of the wall plane is projected onto the second preset plane to obtain the wall surface line segment of the wall plane;

[0106] The spatial structure information of the target house is determined by using the wall line segments of the wall plane and the second outer rectangular outline.

[0107] As an optional embodiment of this disclosure, the spatial structure determination subunit determines the spatial structure information of the target house using the wall surface segments of the wall plane and the second outer rectangular outline, including:

[0108] Based on the position coordinates of the wall line segments on the wall plane, the wall line segments are added to the corresponding coordinate positions of the second outer rectangular outline to obtain the initial spatial structure of the target house.

[0109] Each wall segment in the initial spatial structure of the target house is supplemented with additional wall segments to obtain the supplemented wall segments.

[0110] The final spatial structure formed by the supplemented wall segments and the second outer rectangular outline is determined as the spatial structure information of the target house.

[0111] As an optional embodiment of this disclosure, the second determining unit includes:

[0112] The sub-unit for determining the image to be segmented is used to determine the corresponding image to be segmented based on the spatial structure information of the target house. The vertices of the image to be segmented are the polygons formed by the spatial structure information of the target house, and the edges of the image to be segmented include the common edges of adjacent polygons formed by the spatial structure information of the target house and the edges formed by connecting the vertices of the image to be segmented with the source and sink points of the graph cut, respectively.

[0113] The weight determination subunit is used to determine the weight of each edge in the graph to be segmented based on the vertices, edges, source and sink vertices of the graph cuts.

[0114] The minimum cut determination subunit is used to calculate and determine the minimum cut of the graph to be segmented based on the weight of each edge in the graph to be segmented using a preset graph cut algorithm, so as to obtain the internal and external boundaries of the spatial structure of the target house.

[0115] The connecting subunit is used to connect the internal and external boundaries of the spatial structure of the target house to obtain the first house layout diagram of the target house.

[0116] As an optional embodiment of this disclosure, the connecting subunit connects the internal and external boundaries of the spatial structure of the target house to obtain a first floor plan of the target house, including:

[0117] The polygonal outline of the target house is obtained by concatenating the edges of the internal and external boundaries of the spatial structure of the target house.

[0118] The first target merging vertex is determined from any two vertices of the polygonal contour that meet the first preset merging condition, or the second target merging vertex is determined from multiple vertices of the polygonal contour that meet the second preset merging condition.

[0119] Merging the first target vertices or the second target vertices is performed to obtain a merged polygonal outline, which is the first house layout diagram of the target house.

[0120] As an optional embodiment of this disclosure, the third determining unit includes:

[0121] The contour point cloud acquisition subunit is used to acquire point cloud data information corresponding to the edges of each contour in the first house floor plan based on the first house floor plan of the target house.

[0122] The third projection subunit is used to project the point cloud data information corresponding to the edge of each contour onto the third preset plane and onto the third preset grid to obtain the third occupied grid map. The third preset plane is the plane formed by the vertical axis and the vertical axis in the three-dimensional coordinate system of the point cloud.

[0123] A void region determination sub-unit is used to determine whether a void region is included in the corresponding third occupancy grid for any edge of a contour.

[0124] A window structure determination subunit is used to determine the rectangular bounding box of the hollow area if it is determined that the hollow area is included, and to obtain the size and coordinate position of the hollow area. The hollow area is the window structure on the side of the corresponding outline in the first house floor plan.

[0125] The form drawing subunit is used to draw the empty area on the edge of the corresponding outline in the first house floor plan according to the size and coordinate position of the empty area, so as to obtain the second house floor plan.

[0126] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0127] Memory, used to store computer program products;

[0128] A processor is configured to execute a computer program product stored in the memory, and when the computer program product is executed, to implement the method described in the first aspect above.

[0129] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the method described in the first aspect above.

[0130] According to a fifth aspect of the present disclosure, a computer program product is provided, including computer program instructions that, when executed by a processor, implement the method described in the first aspect above.

[0131] The technical solution of this disclosure acquires point cloud data of a target house and trajectory information of the acquisition device through a point cloud data acquisition device. Then, based on the point cloud data and trajectory information, a floor plan of the target house is automatically generated and output. This disclosure uses point cloud data to acquire the structural data of the target house, enabling automated drawing of the floor plan. Compared to traditional house measurement, point cloud data can more objectively reflect the structural parameters of the house, without relying on the professionalism and experience of the surveyors, thereby improving the accuracy of house measurement and achieving a more precise floor plan.

[0132] The technical solutions of this disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0133] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0134] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0135] Figure 1 is a flowchart illustrating a method for generating a house floor plan structure diagram according to an embodiment of the present disclosure.

[0136] Figure 2 is a flowchart illustrating a method for generating a house floor plan structure diagram according to another embodiment of the present disclosure.

[0137] Figure 3 is a flowchart illustrating a method for generating a house floor plan structure diagram according to another embodiment of the present disclosure.

[0138] Figure 4 is a schematic diagram showing the effect of the spatial structure of the target house according to an embodiment of the method of this disclosure.

[0139] Figure 5 is a flowchart illustrating a method for generating a house floor plan structure diagram according to another embodiment of the method disclosed herein.

[0140] Figure 6 is one of the schematic diagrams illustrating the determination of the inner and outer contour information of a target house using a graph cut algorithm according to an embodiment of the method disclosed herein.

[0141] Figure 7 is a second schematic diagram of a method according to the present disclosure, which uses a graph cut algorithm to determine the inner and outer contour information of a target house.

[0142] Figure 8 is a schematic diagram (3) of an embodiment of the method of this disclosure, which uses a graph cut algorithm to determine the inner and outer contour information of a target house.

[0143] Figure 9 is a flowchart illustrating a method for generating a house floor plan structure diagram according to another embodiment of the method disclosed herein.

[0144] Figure 10 is a schematic diagram of determining the window structure according to an embodiment of the method of this disclosure.

[0145] Figure 11 is a schematic diagram of the final house layout of an embodiment of the method of this disclosure.

[0146] Figure 12 is a structural block diagram of a house floor plan generation device according to an embodiment of the present disclosure.

[0147] Figure 13 is a structural block diagram of a house floor plan generation device according to another embodiment of the present disclosure.

[0148] Figure 14 is a structural block diagram of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0149] In this embodiment, an image acquisition device such as a lidar can be used to perform a 360-degree scan of the building space, acquiring point cloud data of surfaces such as walls, ceilings, and floors, thereby obtaining a point cloud image of the building space. It is understood that the point cloud image of the building space consists of a large set of discrete points, and these discrete points do not have topological relationships, but each point contains spatial coordinates and possibly other attribute information. Therefore, this embodiment utilizes the point cloud data of the building space to obtain the structural data of the target building, enabling automated drawing of the building's floor plan.

[0150] To enable those skilled in the art to clearly understand the technical solutions of the embodiments of this disclosure, the technical solutions of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings.

[0151] Figure 1 is a flowchart illustrating a method for generating a house floor plan according to an embodiment of the present disclosure. As shown in Figure 1, the method for generating a house floor plan according to an embodiment of the present disclosure may include the following steps:

[0152] Step 101: Obtain the first point cloud data information of the target house and the trajectory information of the first point cloud data acquisition device.

[0153] In this embodiment of the disclosure, an image acquisition device such as a LiDAR can be used to perform a 360-degree scan of the plane of each space of the target house, such as the ceiling, floor, and walls, to obtain point cloud data information of the target house. It is understood that in addition to its own structure (ceiling, floor, and walls), the target house may also include auxiliary structures such as doors, wardrobes, cabinets, refrigerators, and interior mirrors. Therefore, the point cloud data information will also include point cloud data of these auxiliary structures.

[0154] In this embodiment of the disclosure, in order to collect point cloud data information of the target house in all directions, the acquisition device can be moved to traverse every space of the target house and all planes of each space to obtain more complete point cloud data information of the target house.

[0155] Step 102: Based on the first point cloud data information and trajectory information, determine the house layout of the target house.

[0156] The point cloud data of the target house consists of a large set of discrete points, each containing spatial coordinates and possible other attribute information. Therefore, in this embodiment of the disclosure, the spatial coordinates of the point cloud data of each space of the target house are used to obtain the spatial structure data of each space, and the trajectory information of the acquisition device can assist in the division of the spatial structure data of each space of the target house, ultimately realizing the automated drawing of the house floor plan.

[0157] A house floor plan can be a CAD drawing, which includes the structural division of each space in the target house, the ancillary structures such as windows and doors, and information such as the proportions and dimensions of each space and ancillary structures.

[0158] Step 103: Output the floor plan of the target house.

[0159] After completing the floor plan of the target house, the floor plan can be output. For example, it can receive output instructions from the input device and output the floor plan in a specific file format, or it can be displayed in CAD drawing software.

[0160] The technical solution of this disclosure acquires point cloud data of a target house and trajectory information of the acquisition device through a point cloud data acquisition device. Then, based on the point cloud data and trajectory information, a floor plan of the target house is automatically generated and output. This disclosure uses point cloud data to acquire the structural data of the target house, enabling automated drawing of the floor plan. Compared to traditional house measurement, point cloud data can more objectively reflect the structural parameters of the house, without relying on the professionalism and experience of the surveyors, thus improving the accuracy of house measurement and achieving a more precise floor plan.

[0161] Based on the embodiment shown in Figure 1, this disclosure also provides another embodiment. Figure 2 is a flowchart illustrating a method for generating a house floor plan according to another embodiment of the method of this disclosure.

[0162] As shown in Figure 2, a method for generating a house floor plan according to an embodiment of this disclosure may further include the following steps:

[0163] Step 101: Obtain the first point cloud data information of the target house and the trajectory information of the first point cloud data acquisition device.

[0164] Please refer to step 101 as shown in Figure 1. For the sake of brevity, it will not be described again here.

[0165] Step 1021: Denoise the first point cloud data information to obtain the second point cloud data information.

[0166] During the point cloud data acquisition process, the living environment inside the house is quite complex. For example, furniture and appliances such as cabinets and refrigerators can block the walls, and reflective objects such as mirrors can interfere with the acquisition equipment's acquisition of point cloud data. Objects seen through windows outside may also be captured, and there may even be people walking around inside the room. All of these factors can cause the raw point cloud data (first point cloud data) acquired by the acquisition equipment to be very messy.

[0167] Based on this, the present disclosure provides a scheme for denoising point cloud data information to obtain cleaner point cloud data information, reduce interference from noisy point cloud data, and improve the accuracy of house floor plan diagrams.

[0168] As an embodiment of this disclosure, a preset point cloud dynamic object removal algorithm can be used to remove the first noisy point cloud from the first point cloud data information to obtain the second point cloud data information. The first noisy point cloud is dynamic object noise information. For example, the preset point cloud dynamic object removal algorithm may use a multi-resolution range image (distance image or depth image) to identify dynamic objects, and use the visibility of the projected range image to identify moving points, thereby dividing the first point cloud data information into dynamic point clouds and static point clouds. The dynamic point cloud is the point cloud data information of dynamic objects (e.g., people moving around in a room), which is identified as the first noisy point cloud. Deleting the first noisy point cloud yields the second point cloud data information. It is understood that the preset point cloud dynamic object removal algorithm includes, but is not limited to, using a multi-resolution range image to identify dynamic objects, which will not be elaborated here. In this embodiment, the preset point cloud dynamic object removal algorithm is used to identify dynamic objects, thereby removing the point cloud of dynamic objects to obtain cleaner and more accurate point cloud data information of the target house's interior space, reducing the impact of noisy point cloud data and improving the accuracy of the house's floor plan.

[0169] As another embodiment of this disclosure, noise reduction can also be applied to the point cloud data emitted by the mirror plane. For example, the wall in the mirror reflects point cloud data, which is noisy point cloud and requires noise reduction processing. Specifically, the coordinate range of the mirror plane can be determined from the first point cloud data, and the point cloud data within the coordinate range of the mirror plane in the first point cloud data can be determined as the second noisy point cloud. The second noisy point cloud is then deleted from the first point cloud data to obtain the second point cloud data. For example, the high-intensity echo characteristics when a laser is perpendicularly incident on a mirror can be used to determine all mirror planes in the target house. Then, the range of the mirror can be recovered using the point cloud of the mirror frame (or the plane point cloud of the mirror boundary if it is a frameless mirror attached to a wall or cabinet). All point cloud data collected by light passing through the area within the mirror can be deleted.

[0170] As another embodiment of this disclosure, since the target house has glass windows and related structures, objects outside the target house can be seen through the glass windows. These objects may also be collected by the acquisition device, but they do not belong to the target house. Therefore, it is necessary to denoise the point cloud data information of these objects. Specifically, the first point cloud data information is projected onto a first preset plane and then onto a preset grid to obtain a first occupancy grid map. The first preset plane is the plane formed by the horizontal and vertical axes in the three-dimensional coordinate system of the point cloud. The trajectory information of the acquisition device for the first point cloud data is projected onto the first preset plane to obtain the two-dimensional trajectory information of the acquisition device on the first preset plane. All connected components formed by the first point cloud data information are determined according to the first occupancy grid map. Based on the two-dimensional trajectory information of the acquisition device on the first preset plane, noisy connected components are determined from all connected components. Then, the point cloud information in the first point cloud data that is in the noisy connected component is determined as the third noisy point cloud. The third noisy point cloud is deleted from the first point cloud data information to obtain the second point cloud data information. In this embodiment, the point cloud data information of the target house is projected onto a grid map, which can group point clouds with similar coordinate positions together (e.g., in one grid) to obtain the main part of the interior of the target house. In addition, since the acquisition device acquires data inside the target house, the area traversed by the two-dimensional trajectory information it forms is the interior, thereby more accurately determining the non-indoor parts, i.e., the non-indoor connected domains, and deleting the point cloud data information in the non-indoor connected domains.

[0171] In a point cloud coordinate system, the ground is not necessarily horizontal. By statistically analyzing the normal distribution of the point cloud, the angle of the point cloud of the target house can be adjusted, for example, rotated to be parallel to the ground. Specifically, before projecting the first point cloud data information onto the first preset plane and then onto the first preset grid to obtain the first occupancy grid map, the normal of the first point cloud data information is adjusted according to the first average normal distribution to obtain the second average normal distribution of the first point cloud data information, wherein the second normal distribution of the first point cloud data information is perpendicular to the first preset plane. Then, the point cloud of the target house is projected onto the horizontal plane to form the occupancy grid map, and the trajectory position is also projected onto the horizontal plane to form two-dimensional trajectory information. Then, each connected component is extracted on the occupancy grid map (i.e., the first occupancy grid map), and connected components without two-dimensional trajectory information are filtered out. In this way, point cloud data of the exterior of the target house collected through glass windows, etc., is removed to obtain more accurate point cloud data information of the target house.

[0172] The average normal distribution of the first point cloud data refers to the average value of the normal vector of each point in the point cloud, used to describe the overall directional distribution characteristics of the point cloud. The statistical calculation of the average normal distribution can be achieved, for example, through least squares plane fitting. For the average normal distribution of cloud data on a plane, it can be expressed using normal vectors as ax + by + cz = d, where a 2 +b 2 +c 2 =1, construct the cost function using the least squares method to make the plane most accurate, and then estimate the average normal distribution of the plane surface by solving the function.

[0173] As another embodiment of this disclosure, the interior of the walls of the target house may also contain noise point clouds caused by various reasons. In this embodiment, the internal and external structures of a wall structure can be determined first, and then the internal structure of the wall can be determined. The point clouds appearing in the internal structure are then denoised. Specifically, two planes of any wall of the target house can be determined based on the first point cloud data information. The point cloud data information in the first point cloud data information that lies between the two planes of any wall of the target house is determined as the fourth noise point cloud. The fourth noise point cloud is then deleted from the first point cloud data information to obtain the second point cloud data information. For example, point cloud data information on both sides of the wall can be used. For instance, if the normals of two planes are opposite and the positions reflected by the point cloud data are close, then these two planes are determined as two planes of a wall. If point clouds appear inside these two planes, then this part of the point cloud data is determined as noise point clouds and denoised.

[0174] Step 1022: Determine the spatial structure information of the target house based on the second point cloud data information.

[0175] Spatial construction information includes at least the maximum spatial range that the point cloud data of the target house can map, and the spatial structure of the target house (e.g., the number of walls, the location of the walls, and the various spaces composed of the walls).

[0176] As an embodiment of this disclosure, determining the spatial structure information of the target house can be achieved through the following steps. Figure 3 is a flowchart illustrating a method for generating a house floor plan according to another embodiment of the method of this disclosure. As shown in Figure 3, determining the spatial structure information of the target house may include the following steps:

[0177] Step 1022a: Determine the wall plane of the target house based on the second point cloud data information.

[0178] Specifically, firstly, the region growing method and the region averaging method can be used to determine all spatial planes of the target house based on the first point cloud data information.

[0179] In this embodiment, the average normal of a region (i.e., the average normal of a planar region of a wall) is combined with a region growing method to determine the spatial plane (e.g., including but not limited to the plane of a wall, floor, ceiling, etc.). Specifically, when using the region growing method for region growth, in addition to determining whether the angle between the point cloud normal and the normal of neighboring points meets a preset angle threshold, it is also necessary to determine whether the angle between the point cloud normal and the average normal of the overall planar region of a wall meets a preset angle threshold (e.g., the preset angle range is between 5° and 10°, including 5° and 10°). In this way, it avoids the region growing method only considering local planarity. For example, when there is noise in the point cloud, the angle threshold between the point cloud normal and the normal of neighboring points is difficult to adjust. If the angle threshold is strict, some large planes will be grown into multiple small fragmented planes. If the angle threshold is relaxed, different planes are easily grown together in curved areas. Therefore, this embodiment can more robustly extract all spatial planes of the target house completely, improving data accuracy.

[0180] Then, the vertical plane is determined from all the spatial planes of the target house using the preset plane equation.

[0181] It is understandable that not all spatial planes of the target house are vertical; for example, there are floor and ceiling planes. Therefore, it is necessary to determine the wall planes from these spatial planes. Specifically, after extracting all spatial planes, we can determine which planes are vertical planes, i.e., spatial planes that satisfy the constraints of the plane equation, by restricting their plane equations (e.g., ax + by + cz = 0, and adding constraints to this equation (e.g., the normal vector of a vertical plane is parallel to the z-axis, i.e., the x and y components of the normal vector are 0, and the z component is not 0), to ensure that they are vertical planes).

[0182] Next, the second point cloud data information is projected onto the second preset plane and then onto the second preset grid to obtain the second occupied grid map, wherein the second preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud.

[0183] Next, delete the vertical planes located inside the second occupied raster to obtain the vertical planes located at the edge of the second occupied raster.

[0184] Finally, the vertical plane located at the edge of the second occupied grid is determined as the wall plane of the target house.

[0185] The vertical plane includes not only wall planes but also planes of large appliances such as refrigerators and door panels. Therefore, it is necessary to distinguish between wall planes and non-wall planes. In this embodiment, it is assumed that, compared to non-wall planes, wall planes are closer to the edge of the grid in the point cloud occupancy grid map, while non-wall planes are closer to the inside of the grid. Based on this, the top-view projection occupancy grid map of the point cloud (i.e., top-view projection onto a second preset grid to obtain a second occupancy grid map) is used to further determine which vertical planes are located inside the second preset grid and which are located at the edge of the second preset grid. In this way, vertical planes completely located inside the space are eliminated, and the remaining vertical planes located at the grid edge are wall planes.

[0186] Step 1022b: Determine the first outer rectangular outline based on the bounding box of the second point cloud data information.

[0187] The bounding box of the second point cloud data information, i.e., the bounding box of the point cloud, is a geometric shape used to approximately represent the boundary range of the point cloud data, mainly for rapid analysis and processing of point cloud data. Common point cloud bounding box types include Axis-Aligned Bounding Box (AABB) and Oriented Bounding Box (OBB). AABB (Axis-Aligned Bounding Box) is an axis-aligned bounding box whose boundaries are parallel to the coordinate axes. AABB calculation is relatively simple, but it may not perfectly fit the shape of the point cloud. OBB (Oriented Bounding Box) can be rotated according to the orientation of the point cloud to more closely fit its shape. OBB calculation is relatively complex, but it provides more accurate boundary information. This disclosure embodiment uses the bounding box of the point cloud data information to quickly determine the boundary range of the point cloud data, improving processing efficiency.

[0188] The first outer rectangular outline is the rectangular outline formed by the boundary range corresponding to the point cloud data information of the target house.

[0189] Step 1022c: Determine the spatial structure information of the target house using the wall plane and the first outer rectangular outline.

[0190] Specifically, firstly, the outline size of the first outer rectangular outline can be enlarged and adjusted to obtain the second outer rectangular outline.

[0191] For example, based on the first outer rectangular outline, the first outer rectangular outline can be enlarged and adjusted according to a preset size to obtain the second outer rectangular outline. The preset size is, for example, enlarged by 5 to 10 mm according to a scale bar, or enlarged to 1 m according to the actual size, etc., and this disclosure does not limit it.

[0192] Then, the point cloud data of the wall plane is projected from the top view to obtain the wall surface line segments.

[0193] For example, projecting any extracted wall plane onto a second preset plane (or top view) yields a straight line. Then, projecting the point cloud data from the first point cloud of the wall plane onto the second preset plane (or top view) onto the straight line, converting the three-dimensional coordinates of the point cloud into two-dimensional coordinates, allows us to obtain the two endpoints of the wall plane projection onto the straight line. This determines the range of the wall plane on the top view projection, forming a calculable range on the straight line, obtaining the two endpoints, and ultimately forming the wall line segment corresponding to the wall plane.

[0194] Finally, the spatial structure information of the target house is determined by using the wall line segments of the wall plane and the second outer rectangular outline.

[0195] Furthermore, based on the position coordinates of the wall line segments on the wall plane, the wall line segments can be added to the corresponding coordinate positions of the second outer rectangular outline to obtain the initial spatial structure of the target house. Then, each wall line segment in the initial spatial structure of the target house is supplemented to obtain the supplemented wall line segments. The final spatial structure formed by the supplemented wall line segments and the second outer rectangular outline is determined as the spatial structure information of the target house.

[0196] Specifically, in this embodiment, wall segments are sequentially added to the rectangular area formed by the second outer rectangular outline. It is understood that the wall segments retain their two-dimensional coordinates in the two-dimensional coordinate system, so their corresponding positions can be determined using their two-dimensional coordinates. All wall segments are sequentially added to the corresponding positions in the rectangular area formed by the second outer rectangular outline. If an endpoint of any wall segment fails to intersect with other existing wall segments, the wall segment can be extended until it intersects with other existing wall segments in the space. In this way, with each addition of a wall segment, the space of the rectangular area formed by the second outer rectangular outline is divided into more fragmented spaces, thereby forming the spatial structure of the target house and obtaining the corresponding spatial structure information.

[0197] As an optional embodiment, in addition to the wall planes, extra line segments can be added to the rectangular area formed by the second outer rectangular outline. Due to factors such as objects obstructing the view of the target house, it may be impossible to extract all wall planes, resulting in insufficient division of the spatial structure of the target house. Therefore, in this embodiment, two line segments perpendicular to each wall plane can be added to both ends of the wall line segment corresponding to that wall plane. For example, the outline of the top-view occupancy grid can be used to add two line segments perpendicular to the wall line segment. This method of addition can utilize the boundaries of the ceiling or floor to assist in the addition of wall lines, thereby obtaining a more complete set of wall lines for the target house and providing a more accurate data foundation for the house floor plan.

[0198] In summary, through the division and addition of wall segments, the rectangular area formed by the second outer rectangular outline is ultimately divided into a two-dimensional polygonal grid, which constitutes the spatial structure information of the target house. Referring to Figure 4, which is a schematic diagram of the spatial structure effect of the target house according to an embodiment of the method of this disclosure, the gray lines represent the wall segments extracted and added sequentially to the rectangular area, while the black lines represent wall segments added based on existing wall segments. For example, the segment pointed to by "1" is a wall segment, and the segment perpendicular to the wall segment pointed to by "A" is the added wall segment. By extracting and adding wall segments sequentially to the rectangular area and adding wall segments to existing wall segments, the entire rectangular area formed by the second outer rectangular outline is divided into multiple spaces, thereby forming the spatial structure information of the target house.

[0199] Step 1023: Based on the spatial structure information of the target house, determine the internal and external contour information of the target house to obtain the first house layout diagram of the target house.

[0200] For example, as shown in FIG5, step 1023 may include the following steps:

[0201] Step 1023a: Determine the corresponding segmentation map based on the spatial structure information of the target house.

[0202] Graph cut is the process of dividing a directed graph into two disjoint sets. In this embodiment, the graph to be divided is the directed graph in the graph cut algorithm. Its vertices are the polygons formed by the spatial structure information of the target house, and its edges include the common edges of adjacent polygons formed by the spatial structure information of the target house, as well as the edges formed by connecting the vertices of the graph to be divided to the source point (e.g., S in Figure 6) and the sink point (e.g., T in Figure 6). These vertices and edges constitute the directed graph to be divided. Taking Figure 6 as an example, for instance, the rectangular polygonal region traversed by point A connecting to point i in Figure 6 includes rectangular polygonal regions numbered 1, 2, 8, 9, 10, and 11. The common edges between these rectangular polygonal regions correspond to the edges of the graph, such as the common edges of the rectangular polygonal regions numbered 11 and 12 occupied by the point cloud in Figure 6, and the common edges of the rectangular polygonal regions numbered 5 and 6.

[0203] Step 1023b: Determine the weight of each edge in the graph to be segmented based on the vertices, edges, source and sink vertices of the graph cut.

[0204] Understandably, to find the minimum cut between the source point (e.g., S in Figure 6) and the sink point (e.g., T in Figure 6) of the graph cut, and to obtain the segmentation of the space inside and outside (i.e., the internal and external boundaries of the spatial structure of the target house), it is also necessary to assign appropriate weights to each edge. In this embodiment of the disclosure, the polygon vertices traversed by the trajectory (circles in the polygon numbered "1" in Figure 6) can be strongly connected to the source point S, and the outermost polygon of the rectangular space (polygon numbered "12" in Figure 6) can be strongly connected to T. That is, the edge between S and T has an infinite weight inf, the edge between T and T has a weight of 0, and the weights of the remaining edges are initialized to 0. Furthermore, each point in the point cloud (circles numbered a, b...j in Figure 6) is connected to its emitting point (point A of the rectangle in Figure 6) to form a line of sight. The edge weight between the polygons through which the line of sight passes is increased by w, and the edge weight of the polygon behind each point in the point cloud connected to T is increased. Specifically, taking Figure 6 as an example, a line segment is formed connecting point A in rectangle to any point cloud (points a, b, c, d, e, f, g, h, i, j in Figure 6). The weight of the edge corresponding to the common edge traversed by this line segment is increased by a fixed weight (e.g., w), which is the final weight value of the second target edge. As shown in Figure 6, the rectangular polygonal regions traversed by point A connecting to point i include rectangular polygonal regions numbered 1, 2, 8, 9, 10, and 11. The common edge between these rectangular polygonal regions corresponds to the edge of the graph, and its corresponding weight value is increased by a fixed weight (e.g., w). The edges of the graph to which the point clouds (points a, b, c, d, e, f, g, h, i, j in Figure 6) belong are subject to a reverse fixed weight (e.g., w') from the outside to the inside. Here, the reverse fixed weight (e.g., w') represents the part with the smallest weight between the starting point S and the ending point T in the graph cut algorithm, that is, the position that is easy to break.

[0205] In other embodiments, to make the boundaries of the inner and outer contours of the target house as simple as possible, a smoothing term can be added, that is, the length of the sides between polygons is used as a weight and added to the weight of the graph to encourage the graph cut to take the shortest path when solving the contour.

[0206] Step 1023c: Using a preset graph cut algorithm, the minimum cut of the graph to be segmented is calculated based on the weight of each edge in the graph to be segmented, thereby obtaining the internal and external boundaries of the spatial structure of the target house.

[0207] After determining the directed graph (the graph to be segmented), a graph cut algorithm is further applied to calculate the minimum cut, thereby obtaining the internal and external boundaries of the spatial structure of the target house. For example, this disclosure embodiment can utilize point cloud visibility information. That is, during image capture, if a data acquisition device such as a LiDAR can capture any point, it can be determined that the line connecting the location of the acquisition device and that point (point cloud) is definitely unobstructed. Each captured point cloud will have a location of the acquisition device. Using this information, in the top view, the polygonal region traversed by the line connecting the location of the acquisition device and that point is determined as the internal boundary of the spatial structure of the target house, and the untraversed polygonal region is determined as the external boundary of the spatial structure of the target house.

[0208] Step 1023d: Connect the internal and external boundaries of the spatial structure of the target house to obtain the first house layout diagram of the target house.

[0209] By setting the weights in the solution space, the spatial structure of the target house is ultimately divided into internal and external boundaries. The edges of the rectangular polygonal regions between the internal and external boundaries are then connected sequentially to form a closed polygonal outline, which is the first house layout diagram of the target house.

[0210] See Figure 7, which is the second schematic diagram of determining the inner and outer contour information of the target house using the graph cut algorithm, i.e., the schematic diagram of the result after the graph cut solution. In Figure 7, the irregular path formed by the circles is the trajectory of the acquisition device, i.e. the position of the emission point of the point cloud. The white line represents the interior, and the gray line represents the exterior. The intersection of the two regions forms the polygonal outline of the target house, which is the first house type structure diagram of the target house.

[0211] However, the polygon contour obtained by directly solving is relatively fragmented and irregular. Therefore, this disclosure provides a technical solution to further optimize the polygon contour to obtain a simpler and more regular polygon contour.

[0212] For example, as one possible implementation of this disclosure, step 1023d can also be implemented as follows: First, the edges of the internal and external boundaries of the spatial structure of the target house are concatenated to obtain the polygonal outline of the target house.

[0213] The first step is to obtain the initial polygon outline, as shown in Figure 8A, which is the initial polygon outline before optimization. It includes many small points and lines, which may cause the entire polygon area to be fragmented and irregular.

[0214] Then, any two vertices of the polygonal contour that meet the first preset merging condition are determined as the first target merging vertices, or the second target merging vertices are determined from multiple vertices of the polygonal contour that meet the second preset merging condition.

[0215] As an optional embodiment, the first preset merging condition aims to merge two nearby vertices. For example, if the distance between two vertices is less than a certain threshold (e.g., 5cm), then the two vertices are determined as the first target vertices to be merged.

[0216] As another alternative embodiment, the aim is to remove some vertices using collinear vertices. A first preset merging condition is, for example, that if multiple consecutive vertices lie on the same straight line, then the intermediate vertex between the first and last vertices is determined as the second target merging vertex.

[0217] Finally, the first target vertices or the second target vertices are merged to obtain the merged polygonal outline, which is the first house layout diagram of the target house.

[0218] For two vertices that satisfy the first objective of merging vertices, for example, two vertices can be merged into one. In this way, the loop will eventually reduce the number of vertices in the polygon outline.

[0219] For multiple vertices that satisfy the first objective of merging vertices and the second objective of merging vertices, deletion can be performed, ultimately retaining the first and last vertices on a certain straight line.

[0220] By merging and optimizing the vertices in the polygonal contour as described above, the number of vertices in the polygonal contour can be greatly reduced, making the entire polygonal contour look more regular in terms of line segments and space. See Figure B in Figure 8 for the optimized polygonal contour.

[0221] In this embodiment of the disclosure, after vertex merging optimization, the vertex positions can be further optimized and adjusted so that the approximately horizontal or vertical line segments are strictly horizontal and vertical, the approximately parallel or perpendicular line segments remain parallel or perpendicular, and the line segment positions remain near their respective wall support point clouds.

[0222] Step 1024: Based on the first floor plan of the target house, determine the window structure of the target house to obtain the second floor plan. The second floor plan is the final floor plan of the target house.

[0223] The inner plane of the wall plane will form a hole at the position of the window (window structure), that is, there is no point cloud data information. In this embodiment of the disclosure, the spatial information of the window can be recovered by finding the point cloud hole on each wall plane, and then the window can be added to the first house floor plan to obtain a complete house system, that is, the second house floor plan.

[0224] For example, step 1024 can be implemented as shown in Figure 9. As shown in Figure 9, it specifically includes the following steps:

[0225] Step 1024a: Obtain the point cloud data information corresponding to the edges of each contour in the first house layout diagram.

[0226] Step 1024b: Project the point cloud data information corresponding to the edge of each contour onto the third preset plane and onto the third preset grid to obtain the third occupied grid map, wherein the third preset plane is the plane formed by the vertical axis and the vertical axis in the point cloud three-dimensional coordinate system.

[0227] Step 1024c: For any contour edge, determine whether it includes a hole region in its corresponding third occupancy grid.

[0228] As shown in Figure 10, region G in the third occupancy grid diagram represents the hole region.

[0229] Step 1024d: If it is determined that there is a void area, then determine the rectangular bounding box of the void area and obtain the size and coordinate position of the void area. The void area is the edge window structure of the corresponding outline in the first house floor plan.

[0230] As shown in Figure 10, the F region corresponding to the rectangle in Figure B is the bounding box of the hole region, which contains information such as the size and coordinate position of the hole region.

[0231] Step 1024e: Draw the empty area on the edge of the corresponding outline in the first house floor plan according to the size and coordinate position of the empty area, and obtain the second house floor plan.

[0232] For example, in this embodiment of the disclosure, point clouds near each edge of the optimized polygonal contour can be extracted, and the point clouds can be projected onto a side view of the wall plane to obtain a side view occupancy grid map (i.e., a third occupancy grid map). Rectangular bounding boxes of empty areas are extracted from the third occupancy grid map as window structures. The horizontal range of the rectangular bounding box in the empty area corresponds to the width of the window structure, and the vertical range corresponds to the height of the window structure. Further, based on the width and height of the window structure, a conversion between the three-dimensional and two-dimensional coordinates of the point cloud can be performed to obtain the coordinate position and size range of the window structure on the edge of the contour in the first house floor plan, and this is then drawn to obtain a first house floor plan including the window structure, i.e., a second house floor plan.

[0233] Step 103: Output the floor plan of the target house.

[0234] In some embodiments of this disclosure, based on the assumption of a fixed thickness of the wall structure of the target house (for example, in drawing, the thickness of the wall structure is generally 20cm to 40cm), the outline of the inner wall plane of the wall structure can be uniformly expanded outward by a certain range (for example, expanded by 20cm to 40cm) to obtain the position and outline of the outer wall plane of the wall structure, so that the house floor plan of the target house includes the outlines of both the inner and outer wall planes. Furthermore, the two endpoints of the line segment corresponding to the window structure are extended vertically outward to intersect with the outline of the outer wall plane of the wall structure. This makes the wall structure, spatial structure, etc., of the house floor plan of the target house more aesthetically pleasing. See Figure 11 for a specific example of the house floor plan. The four positions A, B, C, and D in the figure indicate the positions of the window structures of the target house. The inner outline in the figure represents the outline of the inner wall plane of the wall structure, and the outer outline represents the outline of the outer wall plane of the wall structure.

[0235] The technical solution of this disclosure acquires point cloud data of a target house and trajectory information of the acquisition device through a point cloud data acquisition device. Then, based on the point cloud data and trajectory information, a final floor plan of the target house is automatically generated and output. This disclosure uses point cloud data to acquire the structural data of the target house, enabling automated drawing of the floor plan. Compared to traditional house measurement, point cloud data can more objectively reflect the structural parameters of the house, without relying on the professionalism and experience of the surveyors, thereby improving the accuracy of house measurement and achieving a more precise floor plan.

[0236] Correspondingly, this disclosure also provides apparatus embodiments corresponding to the foregoing method embodiments. Figure 12 is a structural block diagram of an apparatus for generating a house floor plan according to an embodiment of the apparatus of this disclosure. As shown in Figure 12, an apparatus for generating a house floor plan according to an embodiment of the apparatus of this disclosure may include a data acquisition module 1201, a house floor plan determination module 1202, and a house floor plan output module 1203. Wherein:

[0237] The data acquisition module 1201 is used to acquire the first point cloud data information of the target house and the trajectory information of the acquisition device of the first point cloud data;

[0238] The house layout diagram determination module 1202 is used to determine the house layout diagram of the target house based on the first point cloud data information and trajectory information;

[0239] The house floor plan output module 1203 is used to output the house floor plan of the target house.

[0240] The technical solution of this disclosure acquires point cloud data of a target house and trajectory information of the acquisition device through a point cloud data acquisition device. Then, based on the point cloud data and trajectory information, a floor plan of the target house is automatically generated and output. This disclosure uses point cloud data to acquire the structural data of the target house, enabling automated drawing of the floor plan. Compared to traditional house measurement, point cloud data can more objectively reflect the structural parameters of the house, without relying on the professionalism and experience of the surveyors, thereby improving the accuracy of house measurement and achieving a more precise floor plan.

[0241] Based on the embodiment shown in Figure 12, another device embodiment is also provided. Figure 13 is a structural block diagram of a house floor plan generation device according to another embodiment of the present disclosure. As shown in Figure 13, the house floor plan generation device according to an embodiment of the present disclosure may further include:

[0242] The house layout diagram determination module 1202 may include: a noise reduction processing unit 1202a, a first determination unit 1202b, a second determination unit 1202c, and a third determination unit 1202d. Wherein:

[0243] The denoising processing unit 1202a is used to denoise the first point cloud data information to obtain the second point cloud data information.

[0244] As an optional embodiment, the noise reduction processing unit 1202a may further include:

[0245] The first noise removal subunit is used to remove the first noise point cloud from the first point cloud data information using a preset point cloud dynamic object removal algorithm to obtain the second point cloud data information, wherein the first noise point cloud is dynamic object noise information.

[0246] As another optional embodiment, the noise reduction processing unit 1202a may further include:

[0247] The mirror plane determination subunit is used to determine the coordinate range of the mirror plane from the first point cloud data information;

[0248] The first noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information that is within the coordinate range of the mirror plane as the second noise point cloud.

[0249] The second noise removal subunit deletes the second noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0250] As another optional embodiment, the noise reduction processing unit 1202a may further include:

[0251] The first projection subunit projects the first point cloud data information onto a first preset plane onto a preset grid to obtain a first occupied grid map, wherein the first preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud;

[0252] The second projection subunit is used to project the trajectory information of the acquisition device of the first point cloud data onto the first preset plane to obtain the two-dimensional trajectory information of the acquisition device on the first preset plane.

[0253] The connected component determination subunit is used to determine all connected components formed by the first point cloud data information based on the first occupancy raster map.

[0254] The noise connectivity determination subunit is used to determine the noise connectivity from all the connectivity based on the two-dimensional trajectory information of the acquisition device on the first preset plane.

[0255] The second noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information and in the noise connected domain as the third noise point cloud.

[0256] The third noise removal subunit is used to delete the third noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0257] As another optional embodiment, the noise reduction processing unit 1202a may further include:

[0258] The second projection subunit is used to adjust the normal of the first point cloud data information according to the first average normal distribution of the first point cloud data information before projecting the first point cloud data information onto the first preset plane onto the first preset grid to obtain the first occupied grid map, thereby obtaining the second average normal distribution of the first point cloud data information, wherein the second average normal distribution of the first point cloud data information is perpendicular to the first preset plane.

[0259] As another optional embodiment, the noise reduction processing unit 1202a may further include:

[0260] The first wall surface determination subunit is used to determine two planes of any wall of the target house based on the first point cloud data information.

[0261] The third noise point cloud determination unit is used to determine the point cloud data information in the first point cloud data information that is between two planes on any wall of the target house as the fourth noise point cloud;

[0262] The fourth noise removal subunit is used to delete the fourth noise point cloud from the first point cloud data information to obtain the second point cloud data information.

[0263] The first determining unit 1202b is used to determine the spatial structure information of the target house based on the second point cloud data information.

[0264] As an optional embodiment, the first determining unit 1202b may include:

[0265] The second wall surface determination subunit is used to determine the wall plane of the target house based on the second point cloud data information;

[0266] The rectangular outline determination subunit is used to determine the first outer rectangular outline based on the bounding box of the second point cloud data information.

[0267] The spatial structure determination subunit is used to determine the spatial structure information of the target house using the wall plane and the first outer rectangular outline.

[0268] Optionally, the second wall determination subunit can be further used to: determine all spatial planes of the target house based on the second point cloud data information using the region growing method and the region average normal; determine the vertical plane from all spatial planes of the target house using a preset plane equation; project the second point cloud data information onto a second preset plane into a second preset grid to obtain a second occupancy grid map, wherein the second preset plane is the plane formed by the horizontal axis and the vertical axis in the point cloud three-dimensional coordinate system; delete the vertical plane located inside the second occupancy grid map to obtain the vertical plane located at the edge of the second occupancy grid map; and determine the vertical plane located at the edge of the second occupancy grid map as the wall plane of the target house.

[0269] Optionally, the spatial structure determination subunit can be further used to: enlarge and adjust the outline size of the first outer rectangular outline to obtain a second outer rectangular outline; project the point cloud data information of the wall plane onto the second preset plane to obtain the wall surface line segment of the wall plane; and determine the spatial structure information of the target house using the wall surface line segment of the wall plane and the second outer rectangular outline.

[0270] Furthermore, the spatial structure determination subunit uses the wall segments of the wall plane and the second outer rectangular outline to determine the spatial structure information of the target house. Specifically, this can be achieved by: adding the wall segments to the corresponding coordinate positions of the second outer rectangular outline according to the position coordinates of the wall segments of the wall plane to obtain the initial spatial structure of the target house; supplementing each wall segment in the initial spatial structure of the target house to obtain the supplemented wall segments; and determining the final spatial structure formed by the supplemented wall segments and the second outer rectangular outline as the spatial structure information of the target house.

[0271] The second determining unit 1202c is used to determine the inner and outer contour information of the target house based on the spatial structure information of the target house, so as to obtain the first house layout diagram of the target house;

[0272] As an optional embodiment, the second determining unit 1202c may include:

[0273] The sub-unit for determining the image to be segmented is used to determine the corresponding image to be segmented based on the spatial structure information of the target house. The vertices of the image to be segmented are the polygons formed by the spatial structure information of the target house, and the edges of the image to be segmented include the common edges of adjacent polygons formed by the spatial structure information of the target house and the edges formed by connecting the vertices of the image to be segmented with the source and sink points of the graph cut, respectively.

[0274] The weight determination subunit is used to determine the weight of each edge in the graph to be segmented based on the vertices, edges, source and sink vertices of the graph cuts.

[0275] The minimum cut determination subunit is used to calculate and determine the minimum cut of the graph to be segmented based on the weight of each edge in the graph to be segmented using a preset graph cut algorithm, so as to obtain the internal and external boundaries of the spatial structure of the target house.

[0276] The connecting subunit is used to connect the internal and external boundaries of the spatial structure of the target house to obtain the first house layout diagram of the target house.

[0277] Optionally, the connecting subunit concatenates the edges of the internal and external boundaries of the spatial structure of the target house to obtain a first house layout diagram of the target house. Specifically, this can be achieved by: concatenating the edges of the internal and external boundaries of the spatial structure of the target house to obtain a polygonal outline of the target house; determining any two vertices of the image to be segmented in the polygonal outline that satisfy a first preset merging condition as first target merging vertices, or determining second target merging vertices from multiple vertices in the polygonal outline that satisfy a second preset merging condition; and performing a merging process on the first target merging vertices or the second target merging vertices to obtain a merged polygonal outline, which is the first house layout diagram of the target house.

[0278] The third determining unit 1202d is used to determine the window structure of the target house based on the first house layout diagram of the target house, and obtain a second house layout diagram, which is the final house layout diagram of the target house.

[0279] As an optional embodiment, the third determining unit 1202d may include:

[0280] The contour point cloud acquisition subunit is used to acquire point cloud data information corresponding to the edges of each contour in the first house floor plan based on the first house floor plan of the target house.

[0281] The third projection subunit is used to project the point cloud data information corresponding to the edge of each contour onto the third preset plane and onto the third preset grid to obtain the third occupied grid map. The third preset plane is the plane formed by the vertical axis and the vertical axis in the three-dimensional coordinate system of the point cloud.

[0282] A void region determination sub-unit is used to determine whether a void region is included in the corresponding third occupancy grid for any edge of a contour.

[0283] A window structure determination subunit is used to determine the rectangular bounding box of the hollow area if it is determined that the hollow area is included, and to obtain the size and coordinate position of the hollow area. The hollow area is the window structure on the side of the corresponding outline in the first house floor plan.

[0284] The form drawing subunit is used to draw the empty area on the edge of the corresponding outline in the first house floor plan according to the size and coordinate position of the empty area, so as to obtain the second house floor plan.

[0285] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0286] Hereinafter, an electronic device according to an embodiment of the present disclosure will be described with reference to FIG14. The electronic device may be either or both of a first device and a second device, or a standalone device independent of them, which may communicate with the first device and the second device to receive acquired input signals from them.

[0287] Figure 14 illustrates a block diagram of an electronic device according to an embodiment of the present disclosure.

[0288] As shown in Figure 14, the electronic device includes one or more processors and a memory.

[0289] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0290] The memory can store one or more computer program products, and the memory can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program products can be stored on the computer-readable storage medium, and the processor can run the computer program products to implement the methods for generating house floor plan diagrams according to the various embodiments of this disclosure described above, and / or other desired functions.

[0291] In one example, the electronic device may also include input devices and output devices, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0292] In addition, the input device may also include, for example, a keyboard, a mouse, etc.

[0293] This output device can output various information to the outside, including determined distance information, direction information, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0294] Of course, for simplicity, Figure 14 only shows some of the components of the electronic device that are relevant to this disclosure, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the specific application.

[0295] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products, including computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods for generating house floor plan diagrams according to various embodiments of this disclosure as described in the foregoing portion of this specification.

[0296] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this disclosure. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0297] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the method for generating a house floor plan according to various embodiments of this disclosure as described in the foregoing portion of this specification.

[0298] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0299] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0300] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0301] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0302] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0303] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions to this disclosure.

[0304] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0305] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for generating a house floor plan, characterized in that, The method includes: acquiring first point cloud data information of the target house and trajectory information of the acquisition device for the first point cloud data; determining the house layout diagram of the target house based on the first point cloud data information and the trajectory information; and outputting the house layout diagram of the target house; wherein, determining the house layout diagram of the target house based on the first point cloud data information and the trajectory information includes: determining a corresponding image to be segmented based on the spatial structure information of the target house, wherein the vertices of the image to be segmented are the polygons formed by the spatial structure information of the target house, and the edges of the image to be segmented include the target house. The common edges of adjacent polygons formed by the spatial structure information of the house, and the edges formed by connecting the vertices of the graph to be segmented to the source and sink points of the graph cut, respectively; the weight of each edge in the graph to be segmented is determined based on the vertices, edges, source and sink points of the graph cut; the minimum cut of the graph to be segmented is calculated based on the weight of each edge in the graph to be segmented using a preset graph cut algorithm, thereby obtaining the internal and external boundaries of the spatial structure of the target house; the edges of the internal and external boundaries of the spatial structure of the target house are concatenated to obtain the first house layout diagram of the target house.

2. The method according to claim 1, characterized in that, The step of determining the house layout of the target house based on the first point cloud data information and trajectory information further includes: performing noise reduction processing on the first point cloud data information to obtain second point cloud data information; determining the spatial structure information of the target house based on the second point cloud data information; determining the window structure of the target house based on the first house layout of the target house to obtain a second house layout, wherein the second house layout is the final house layout of the target house.

3. The method according to claim 2, characterized in that, The step of denoising the first point cloud data information to obtain the second point cloud data information includes: removing the first noise point cloud from the first point cloud data information using a preset point cloud dynamic object removal algorithm to obtain the second point cloud data information, wherein the first noise point cloud is dynamic object noise information; and / or, determining the coordinate range of the mirror plane from the first point cloud data information; determining the point cloud data information in the first point cloud data information that is within the coordinate range of the mirror plane as the second noise point cloud; deleting the second noise point cloud from the first point cloud data information to obtain the second point cloud data information; and / or, projecting the first point cloud data information onto a first preset plane onto a preset grid to obtain a first occupancy grid map, wherein the first preset plane is the plane formed by the horizontal axis and the vertical axis in the three-dimensional coordinate system of the point cloud; projecting the trajectory information of the acquisition device of the first point cloud data onto the first preset plane to obtain the two-dimensional trajectory information of the acquisition device on the first preset plane; determining all connected components formed by the first point cloud data information according to the first occupancy grid map; and determining the acquisition device's trajectory information according to the first point cloud data information. The system collects two-dimensional trajectory information of the device on a first preset plane, and determines a noise connected region from all connected regions; it identifies point cloud data information in the first point cloud data information that is in the noise connected region as a third noise point cloud; it deletes the third noise point cloud from the first point cloud data information to obtain a second point cloud data information; and / or, before projecting the first point cloud data information onto the first preset plane onto a first preset grid to obtain a first occupancy grid map, it adjusts the normal of the first point cloud data information according to the first average normal distribution of the first point cloud data information to obtain a second average normal distribution of the first point cloud data information, wherein the second average normal distribution of the first point cloud data information is perpendicular to the first preset plane; and / or, based on the first point cloud data information, it determines two planes of any wall of the target house; it identifies point cloud data information in the first point cloud data information that is between the two planes of any wall of the target house as a fourth noise point cloud; it deletes the fourth noise point cloud from the first point cloud data information to obtain a second point cloud data information.

4. The method according to claim 2, characterized in that, The step of determining the spatial structure information of the target house based on the second point cloud data information includes: determining the wall plane of the target house based on the second point cloud data information; determining the first outer rectangular outline according to the outer bounding box of the second point cloud data information; and determining the spatial structure information of the target house using the wall plane and the first outer rectangular outline.

5. The method according to claim 4, characterized in that, The step of determining the wall plane of the target house based on the second point cloud data information includes: using the region growing method and the region average normal to determine all spatial planes of the target house based on the second point cloud data information; using a preset plane equation to determine the vertical plane from all spatial planes of the target house; projecting the second point cloud data information onto a second preset plane into a second preset grid to obtain a second occupancy grid map, wherein the second preset plane is the plane formed by the horizontal axis and the vertical axis in the point cloud three-dimensional coordinate system; deleting the vertical plane located inside the second occupancy grid map to obtain the vertical plane located at the edge of the second occupancy grid map; and determining the vertical plane located at the edge of the second occupancy grid map as the wall plane of the target house.

6. The method according to claim 4 or 5, characterized in that, The step of determining the spatial structure information of the target house using the wall plane and the first outer rectangular outline includes: enlarging and adjusting the outline size of the first outer rectangular outline to obtain a second outer rectangular outline; projecting the point cloud data information of the wall plane onto a second preset plane to obtain the wall surface line segments of the wall plane; and determining the spatial structure information of the target house using the wall surface line segments of the wall plane and the second outer rectangular outline.

7. The method according to claim 6, characterized in that, The step of determining the spatial structure information of the target house using the wall line segments of the wall plane and the second outer rectangular outline includes: adding the wall line segments to the corresponding coordinate positions of the second outer rectangular outline according to the position coordinates of the wall line segments of the wall plane to obtain the initial spatial structure of the target house; supplementing each wall line segment in the initial spatial structure of the target house to obtain the supplemented wall line segments; and determining the final spatial structure formed by the supplemented wall line segments and the second outer rectangular outline as the spatial structure information of the target house.

8. The method according to claim 1, characterized in that, The step of concatenating the edges of the internal and external boundaries of the spatial structure of the target house to obtain a first house layout diagram of the target house includes: concatenating the edges of the internal and external boundaries of the spatial structure of the target house to obtain a polygonal outline of the target house; determining any two vertices of the image to be segmented in the polygonal outline that satisfy a first preset merging condition as first target merging vertices, or determining a second target merging vertices from multiple vertices in the polygonal outline that satisfy a second preset merging condition; and performing a merging process on the first target merging vertices or the second target merging vertices to obtain a merged polygonal outline, wherein the merged polygonal outline is the first house layout diagram of the target house.

9. The method according to any one of claims 2 and 8, characterized in that, The process of determining the window structure of the target house based on the first house layout diagram to obtain a second house layout diagram, which is the final house layout diagram of the target house, includes: acquiring point cloud data information corresponding to the edges of each contour in the first house layout diagram; projecting the point cloud data information corresponding to the edges of each contour onto a third preset plane and then onto a third preset grid to obtain a third occupancy grid diagram, wherein the third preset plane is the plane formed by the vertical axis and the vertical axis in the point cloud three-dimensional coordinate system; determining whether an empty area is included in the corresponding third occupancy grid diagram for any contour edge; if the empty area is determined to be included, determining the rectangular bounding box of the empty area, obtaining the size and coordinate position of the empty area, wherein the empty area is the window structure on the edge of the corresponding contour in the first house layout diagram; drawing the empty area on the edge of the corresponding contour in the first house layout diagram according to the size and coordinate position of the empty area to obtain the second house layout diagram.

Images