Building thermal insulation material detection system and method based on laser point cloud

Abstract

The invention relates to the technical field of point cloud data processing, and particularly discloses a building thermal insulation material detection system and method based on laser point clouds, and the system comprises a processing module which is used for carrying out the region division of all de-noising points of the point cloud data of a building thermal insulation material, and obtaining all de-noising point sets of the point cloud data of the building thermal insulation material; the calculation module is used for obtaining an alternative value of each de-noising point set of the point cloud data of the building thermal insulation material based on all de-noising point sets of the point cloud data of the building thermal insulation material; and the detection module is used for obtaining a detection result of the building thermal insulation material based on the selection and rejection values of all de-noising point sets of the point cloud data of the building thermal insulation material. According to the method, the detection quality aspect and the detection efficiency aspect are integrated, the denoising points needing to be reserved in the point cloud data of the building thermal insulation material in the laser material detection process are accurately screened out, and the detection efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of point cloud data processing, and in particular to a building thermal insulation material detection system and method based on laser point cloud. Background Art

[0002] Currently, traditional two-dimensional image sensing technologies (such as visible light cameras and infrared thermal imagers) face the following core challenges in inspecting building insulation materials: lack of spatial information, sensitivity to environmental interference, and difficulty in quantifying defects. These issues lead to cumbersome and costly inspections, low efficiency, and a high risk of misjudgment. Three-dimensional laser point cloud sensing, an active three-dimensional sensing method and also a type of image sensing technology (three-dimensional point cloud can be considered a derivative of the broader concept of three-dimensional image sensing, i.e., three-dimensional image sensing technology), directly obtains the three-dimensional coordinates, reflection intensity, and spatial distribution of the target surface by emitting pulses and measuring the reflected signals through LiDAR (an intelligent sensor with a range sensor as its core). This technology offers irreplaceable advantages in building insulation material inspection: full spatial information coverage, strong resistance to environmental interference, high-precision quantitative analysis, and contactless, efficient data collection. However, despite the significant advantages of laser point cloud technology, the processing and analysis of laser point cloud data is inherently complex. As the data volume increases, processing time also increases, resulting in a rapid decline in inspection efficiency and making it difficult to meet the needs of rapid inspection or real-time monitoring. Therefore, how to accurately screen the valid points in the point cloud data of building insulation materials through image detection technology while ensuring the quality of detection, and eliminate noise points to reduce invalid calculations has become a key issue that needs to be solved urgently.

[0003] However, the existing laser point cloud-based building insulation material detection system and method only improves the detection performance of difficult samples with too few internal point clouds or too uneven internal point cloud distribution, and solves the problem that point cloud-based 3D target detection performs poorly at long distances, under occlusion, and at excessive steering angles. It does not consider how to achieve accurate screening of denoised points that need to be retained in the point cloud data of building insulation materials during the laser material detection process in terms of comprehensive detection quality and detection efficiency, so as to improve detection efficiency. For example, the patent publication number is "CN112613450B" and the name of the patent is "A 3D target detection method for enhancing performance on difficult samples", and the method includes the following steps: 1. Voxelize the point cloud and then extract features from these voxels; 2. Use the 3D target detection network to obtain a series of target proposal boxes based on the feature map; 3. Predict the category to which the number of points in each candidate box belongs and the unit distance vector between the center point of the candidate box and the average value of the coordinates of all points in the box, and use the auxiliary loss module for classification and regression; 4. Optimize the loss function, perform foreground and background classification, regression of the center point position, shape and steering angle, and binary classification of the object orientation to obtain the final 3D target detection result. The above patent effectively improves the detection performance for difficult samples with too few internal point clouds and too uneven distribution of internal point clouds, and solves the problem that point cloud-based 3D target detection performs poorly at long distances, occlusions, and overly biased steering angles. However, this patent only improves the detection performance for difficult samples with too few internal point clouds or too uneven distribution of internal point clouds, and solves the problem that point cloud-based 3D target detection performs poorly at long distances, under occlusion, and at excessive steering angles. It does not consider how to achieve accurate screening of denoised points that need to be retained in the point cloud data of building insulation materials during laser material detection in terms of comprehensive detection quality and detection efficiency, so as to improve detection efficiency.

[0004] Therefore, the present invention proposes a building insulation material detection system and method based on laser point cloud, aiming to deeply integrate intelligent sensors, intelligent sensing systems with image detection and image recognition technologies to break through the existing technical bottlenecks. Summary of the Invention

[0005] The present invention provides a building insulation material detection system and method based on laser point cloud, which is used to divide all denoised points of point cloud data of building insulation materials into regions, obtain all denoised point sets of point cloud data of building insulation materials, facilitate subsequent calculation of the cut-off value of each denoised point set of point cloud data of building insulation materials, obtain the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of point cloud data of building insulation materials based on all denoised point sets of point cloud data of building insulation materials, and realize the measurement of the distribution density, spatial distribution integrity and acquisition process difficulty of each denoised point set of point cloud data of building insulation materials. Then, according to the density evaluation value, integrity evaluation value and process evaluation value of all denoised point sets of the point cloud data of the building thermal insulation materials, the cut-off value of each denoised point set of the point cloud data of the building thermal insulation materials is obtained, and the priority of retaining each denoised point set of the point cloud data of the building thermal insulation materials in the laser material detection process is accurately quantified. Finally, according to the cut-off value of all denoised point sets of the point cloud data of the building thermal insulation materials, the detection result of the building thermal insulation materials is obtained. In terms of comprehensive detection quality and detection efficiency, the denoised points that need to be retained in the point cloud data of the building thermal insulation materials in the laser material detection process are accurately screened out, which greatly improves the detection efficiency.

[0006] The present invention provides a building insulation material detection system based on laser point cloud, comprising: An acquisition module is used to perform denoising on the collected point cloud data of the building thermal insulation material to obtain all denoised points of the point cloud data of the building thermal insulation material; a processing module, configured to divide all denoised points of the point cloud data of the building thermal insulation material into regions, and obtain all denoised point sets of the point cloud data of the building thermal insulation material; a calculation module for obtaining, based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, and obtaining, based on the density evaluation value, the integrity evaluation value, and the process evaluation value of all denoised point sets of the point cloud data of the building thermal insulation material, a cut-off value for each denoised point set of the point cloud data of the building thermal insulation material; The detection module is used to obtain the detection result of the building insulation material based on the cut-off value of all denoised point sets of the point cloud data of the building insulation material.

[0007] Preferably, the building insulation material detection system based on laser point cloud, the acquisition module includes: The acquisition submodule is used to obtain all three-dimensional points of the point cloud data of building insulation materials; a preprocessing submodule for obtaining a noise point detection space for each 3D point in the point cloud data of the building thermal insulation material, using each 3D point in the point cloud data of the building thermal insulation material as a sphere center and a preset first distance as a radius, and treating remaining 3D points within the noise point detection space for each 3D point in the point cloud data of the building thermal insulation material as neighboring 3D points of the corresponding 3D point in the point cloud data of the building thermal insulation material; The denoising submodule is used to obtain all denoised points of the point cloud data of the building thermal insulation material based on all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material.

[0008] Preferably, the laser point cloud-based building insulation material detection system, the denoising submodule, includes: a preprocessing unit, configured to treat the distance between each three-dimensional point of the point cloud data of the building thermal insulation material and each neighboring three-dimensional point of the corresponding three-dimensional point as the neighboring distance of each neighboring three-dimensional point of each three-dimensional point of the point cloud data of the building thermal insulation material, treat the mean of the neighboring distances of all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material as the mean of the neighboring distances of each three-dimensional point of the point cloud data of the building thermal insulation material, and treat the sum of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material and the standard deviation of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material as the denoising distance of the point cloud data of the building thermal insulation material; The denoising unit is used to treat all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building insulation material, whose distance from the corresponding three-dimensional point is less than the denoising distance of the point cloud data of the building insulation material, as the denoising points of the point cloud data of the building insulation material.

[0009] Preferably, the building insulation material detection system based on laser point cloud, the processing module includes: a processing submodule, configured to use the position of the building insulation material when the three-dimensional laser scanner scans the building insulation material as the origin of the region division of the point cloud data of the building insulation material; The division submodule is used to use the area division origin of the point cloud data of the building thermal insulation material as the center of the sphere, 1 unit length and 0 unit length as the radius, to obtain the spherical layer area, and use the obtained spherical layer area as the first division area of the point cloud data of the building thermal insulation material, and regard all denoised points located in the first division area of the point cloud data of the building thermal insulation material as a denoised point set of the point cloud data of the building thermal insulation material; use the area division origin of the point cloud data of the building thermal insulation material as the center of the sphere, 2 unit lengths and 1 unit length as the radius, to obtain the spherical layer area, and use the obtained spherical layer area as the second division area of the point cloud data of the building thermal insulation material, and regard all denoised points located in the second division area of the point cloud data of the building thermal insulation material as a denoised point set of the point cloud data of the building thermal insulation material; continue the cycle of the above-mentioned division area determination process until all denoised points of the point cloud data of the building thermal insulation material have been delineated as denoised point sets, and obtain all denoised point sets of the point cloud data of the building thermal insulation material.

[0010] Preferably, the building insulation material detection system based on laser point cloud, the computing module includes: A first evaluation submodule is configured to use the quotient of the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material and the volume of the divided area corresponding to the corresponding denoised point set as the density value of the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the quotient of the density value of each denoised point set of the point cloud data of the building thermal insulation material and the average of the density values of all denoised point sets of the point cloud data of the building thermal insulation material as the density evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; The second evaluation submodule is used to obtain an integrity evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The third evaluation submodule is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The calculation submodule is used to obtain the cut-off value of each denoised point set of the point cloud data of the building insulation material based on the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of the building insulation material.

[0011] Preferably, the second evaluation submodule of the building insulation material detection system based on laser point cloud includes: an assignment unit for obtaining a volume of a cavity region of each denoised point set of the point cloud data of the building thermal insulation material based on each denoised point set of the point cloud data of the building thermal insulation material and a preset cavity detection model; A first evaluation unit is configured to use the reciprocal of a quotient between the volume of a hole region of each denoised point set of the point cloud data of the building thermal insulation material and the volume of a divided region corresponding to the denoised point set as an integrity evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; Among them, the preset void detection model is composed of a large number of denoised point sets and their corresponding divided areas of the pre-collected point cloud data of building insulation materials as model input, and the void area volume of the corresponding denoised point set manually marked and delineated as model output. The trained single denoised point set and its corresponding divided area that can be input into the point cloud data of building insulation materials can output the void area volume of the corresponding denoised point set of the point cloud data of building insulation materials.

[0012] Preferably, the third evaluation submodule of the building thermal insulation material detection system based on laser point cloud includes: An acquisition unit, used for acquiring acquisition time of all denoised points in each denoised point set of point cloud data of building thermal insulation materials; The second evaluation unit is configured to use the average of the acquisition times of all denoised points in all denoised point sets of the point cloud data of the building thermal insulation material as the classification time value of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are less than the classification time value of the point cloud data of the building thermal insulation material as the first-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are not less than the classification time value of the point cloud data of the building thermal insulation material as the second-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material; The third evaluation unit is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all first-category denoised points and all second-category denoised points of each denoised point set of the point cloud data of the building thermal insulation material.

[0013] Preferably, the building insulation material detection system based on laser point cloud, the detection module includes: a rejection threshold determination submodule, for using the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material as the horizontal coordinate and the rejection value of the corresponding denoised point set as the vertical coordinate to obtain all rejection points of the point cloud data of the building thermal insulation material, and using each rejection point of the point cloud data of the building thermal insulation material as a circle point and a preset second distance as a radius to obtain a rejection circle area of each rejection point of the point cloud data of the building thermal insulation material, and using the number of remaining rejection points within the rejection circle area of each rejection point of the point cloud data of the building thermal insulation material as the rejection statistic value of the corresponding rejection point of the point cloud data of the building thermal insulation material, using the rejection point with the largest rejection statistic among all the rejection points of the point cloud data of the building thermal insulation material as the threshold point of the point cloud data of the building thermal insulation material, and using the vertical coordinate of the threshold point of the point cloud data of the building thermal insulation material as the rejection threshold of the point cloud data of the building thermal insulation material; The detection submodule obtains the detection results of the building insulation materials based on the cut-off threshold of the point cloud data of the building insulation materials and the cut-off value of all denoised point sets.

[0014] Preferably, the building insulation material detection system based on laser point cloud, the detection submodule includes: a first detection unit, configured to determine whether a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is greater than a cut-off threshold of the point cloud data of the building thermal insulation material; if so, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a discarded denoised point set of the point cloud data of the building thermal insulation material; The second detection unit is used to re-divide the divided area corresponding to each discarded denoised point set of the point cloud data of the building thermal insulation material, obtain all subsets of each discarded denoised point set of the point cloud data of the building thermal insulation material, and judge whether the cut-off value of each subset of each discarded denoised point set of the point cloud data of the building thermal insulation material is greater than the cut-off threshold of the point cloud data of the building thermal insulation material. If so, the corresponding subset of the discarded denoised point set of the point cloud data of the building thermal insulation material is regarded as the retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the point cloud data of the building thermal insulation material is retained. The corresponding subset of the corresponding discarded denoised point set of the cloud data is used as the discarded denoised point set of the point cloud data of the building thermal insulation material, and the above-mentioned re-division process is continued to be repeated for the divided area corresponding to the discarded denoised point set of the newly obtained point cloud data of the building thermal insulation material until the volume of the divided area corresponding to the discarded denoised point set of the point cloud data of the latest obtained building thermal insulation material is less than the preset stop division volume, and the re-division of the divided area corresponding to the discarded denoised point set of the point cloud data of the building thermal insulation material is stopped, and all the retained denoised point sets of the point cloud data of the building thermal insulation material are obtained; The third detection unit is used to obtain the contour detection result of the building thermal insulation material based on all denoised points of all retained denoised point sets of the point cloud data of the building thermal insulation material, and use the contour detection result of the building thermal insulation material as the detection result of the building thermal insulation material.

[0015] The present invention provides a method for detecting building thermal insulation materials based on laser point cloud, which is applied to implement any one of the building thermal insulation material detection systems based on laser point cloud in Examples 1 to 9, comprising: S1: De-noising the collected point cloud data of the building thermal insulation material to obtain all the de-noised points of the point cloud data of the building thermal insulation material; S2: Divide all denoised points of the point cloud data of the building thermal insulation material into regions to obtain all denoised point sets of the point cloud data of the building thermal insulation material; S3: Based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material are obtained, and based on the density evaluation value, integrity evaluation value, and process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is obtained; S4: Obtaining a detection result of the building thermal insulation material based on the cut-off value of each denoised point set of the point cloud data of the building thermal insulation material.

[0016] The beneficial effects of the present invention compared with the prior art are as follows: all denoised points of the point cloud data of building thermal insulation materials are divided into regions to obtain all denoised point sets of the point cloud data of building thermal insulation materials, which is convenient for subsequent calculation of the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials, and the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of building thermal insulation materials are obtained according to all denoised point sets of the point cloud data of building thermal insulation materials, so as to realize the quantification of the distribution density, spatial distribution integrity and acquisition process difficulty of the denoised points in each denoised point set of the point cloud data of building thermal insulation materials, and then the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of building thermal insulation materials are obtained according to all denoised point sets of the point cloud data of building thermal insulation materials. Density evaluation value, integrity evaluation value and process evaluation value are used to obtain the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials, and the priority of retaining each denoised point set of the point cloud data of building thermal insulation materials in the laser material detection process is accurately quantified. Finally, according to the cut-off value of all denoised point sets of the point cloud data of building thermal insulation materials, the detection results of building thermal insulation materials are obtained. In terms of comprehensive detection quality and detection efficiency, the denoised points that need to be retained in the point cloud data of building thermal insulation materials in the laser material detection process are accurately screened out, which greatly improves the detection efficiency. It deeply integrates intelligent sensors, intelligent sensing systems and image detection and image recognition technologies, and breaks through the existing technical bottlenecks.

[0017] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written application documents.

[0018] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 Schematic diagram of a building thermal insulation material detection system based on laser point cloud in an embodiment of the present invention; Figure 2 This is a flow chart of a method for detecting building thermal insulation materials based on laser point cloud in an embodiment of the present invention.

[0020] Figure 3 This is a detector for a building thermal insulation material detection system based on laser point cloud in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0022] Example 1: The present invention provides a building insulation material detection system based on laser point cloud, referring to Figure 1 ,include: An acquisition module is used to perform denoising on the collected point cloud data of the building thermal insulation material to obtain all denoised points of the point cloud data of the building thermal insulation material; a processing module, configured to divide all denoised points of the point cloud data of the building thermal insulation material into regions, and obtain all denoised point sets of the point cloud data of the building thermal insulation material; a calculation module for obtaining, based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, and obtaining, based on the density evaluation value, the integrity evaluation value, and the process evaluation value of all denoised point sets of the point cloud data of the building thermal insulation material, a cut-off value for each denoised point set of the point cloud data of the building thermal insulation material; The detection module is used to obtain the detection result of the building insulation material based on the cut-off value of all denoised point sets of the point cloud data of the building insulation material.

[0023] In this embodiment, the building insulation material is a building insulation material detection system based on laser point cloud (in this embodiment, the building insulation material detection system based on laser point cloud is a detector, and the specific structure of the detector is referenced). Figure 3 ,like Figure 3 The detector shown includes a display 1 and a placement table 2. The display 1 is installed on the placement table 2. A laser measurement component can be provided in the display 1. The laser measurement component can be set at the inner bottom of the display 1. The laser measurement component is used to obtain point cloud data of building insulation materials. A weighing component can also be provided on the placement table 2. The weighing component is used to obtain the weight of the building insulation material. The positions of the laser measurement component and the weighing component can be adjusted and set according to actual needs. The present invention is not limited to this, as long as it can achieve the corresponding function) for laser contour detection of building insulation materials, such as polystyrene boards.

[0024] In this embodiment, the point cloud data is obtained by a 3D scanning device (such as Figure 3 The laser measurement component in the display 1) collects a set of three-dimensional points (spatial coordinate points) used to describe the geometric shape and surface features of the object.

[0025] In this embodiment, denoising is a process of removing noise points introduced into point cloud data due to factors such as scanning device errors, environmental interference, or object surface characteristics (abnormal three-dimensional points that do not match the geometric shape of the real object surface due to factors such as scanning device errors, environmental interference, or object surface characteristics), thereby improving the quality and usability of point cloud data.

[0026] In this embodiment, the denoised points are all the three-dimensional points in the point cloud data of the building thermal insulation material, excluding the noise points.

[0027] In this embodiment, the region division is a process of dividing the locations of all denoised points of the point cloud data of the building insulation material into regions, and then grouping all denoised points of the point cloud data of the building insulation material (a group is a denoised point set, and a denoised point set is a collection of at least one denoised point).

[0028] In this embodiment, the density evaluation value is a value obtained based on all denoised point sets of the point cloud data of the building thermal insulation material, which can reflect the distribution density of the denoised points in each denoised point set of the point cloud data of the building thermal insulation material.

[0029] In this embodiment, the integrity evaluation value is a numerical value obtained based on all denoised point sets of the point cloud data of the building insulation material, which can reflect the completeness of the spatial distribution of the denoised points in each denoised point set of the point cloud data of the building insulation material.

[0030] In this embodiment, the process evaluation value is a numerical value obtained based on all denoised point sets of the point cloud data of the building insulation material, which can reflect the difficulty of the denoised point acquisition process in each denoised point set of the point cloud data of the building insulation material.

[0031] In this embodiment, the cut-off value is the density evaluation value, integrity evaluation value and process evaluation value of all denoised point sets based on the point cloud data of the building insulation material, which can characterize the priority of retaining each denoised point set of the point cloud data of the building insulation material during the laser material detection process.

[0032] In this embodiment, the detection result of the building insulation material is the contour detection result of the building insulation material (the detection result can be Figure 3 (displayed on the display inside the

[0033] The beneficial effects of the above technology are: all denoised points of the point cloud data of building thermal insulation materials are divided into regions, and all denoised point sets of the point cloud data of building thermal insulation materials are obtained, which is convenient for the subsequent calculation of the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials. According to all denoised point sets of the point cloud data of building thermal insulation materials, the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of building thermal insulation materials are obtained, and the distribution density, spatial distribution integrity and acquisition process difficulty of each denoised point set of the point cloud data of building thermal insulation materials are quantified separately, and then the density of the denoised points in each denoised point set of the point cloud data of building thermal insulation materials are quantified separately. The density evaluation value, integrity evaluation value and process evaluation value of all denoised point sets of the point cloud data of thermal insulation materials are calculated, and the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials is obtained. This realizes the accurate quantification of the priority of retaining each denoised point set of the point cloud data of building thermal insulation materials in the process of laser material detection. Finally, according to the cut-off value of all denoised point sets of the point cloud data of building thermal insulation materials, the detection result of building thermal insulation materials is obtained. In terms of comprehensive detection quality and detection efficiency, the denoised points that need to be retained in the point cloud data of building thermal insulation materials in the process of laser material detection are accurately screened out, which greatly improves the detection efficiency.

[0034] Example 2: Based on Example 1, a building insulation material detection system based on laser point cloud, the acquisition module includes: The acquisition submodule is used to obtain all three-dimensional points of the point cloud data of building insulation materials; a preprocessing submodule for obtaining a noise point detection space for each 3D point in the point cloud data of the building thermal insulation material, using each 3D point in the point cloud data of the building thermal insulation material as a sphere center and a preset first distance as a radius, and treating remaining 3D points within the noise point detection space for each 3D point in the point cloud data of the building thermal insulation material as neighboring 3D points of the corresponding 3D point in the point cloud data of the building thermal insulation material; The denoising submodule is used to obtain all denoised points of the point cloud data of the building thermal insulation material based on all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material.

[0035] In this embodiment, the preset first distance is a preset distance of a noise point detection space for each three-dimensional point of the point cloud data of the building thermal insulation material, for example, 5 cm.

[0036] In this embodiment, the noise detection space of each three-dimensional point of the point cloud data of the building insulation material is a spherical space obtained with each three-dimensional point of the point cloud data of the building insulation material as the center and a preset first distance as the radius.

[0037] The beneficial effects of the above technology are: according to each three-dimensional point of the point cloud data of the building insulation material, the noise detection space of each three-dimensional point of the point cloud data of the building insulation material is obtained, and then all the neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building insulation material are obtained, which facilitates the subsequent acquisition of all denoised points of the point cloud data of the building insulation material.

[0038] Example 3: Based on Example 2, a building insulation material detection system based on laser point cloud, a denoising submodule, includes: a preprocessing unit, configured to treat the distance between each three-dimensional point of the point cloud data of the building thermal insulation material and each neighboring three-dimensional point of the corresponding three-dimensional point as the neighboring distance of each neighboring three-dimensional point of each three-dimensional point of the point cloud data of the building thermal insulation material, treat the mean of the neighboring distances of all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material as the mean of the neighboring distances of each three-dimensional point of the point cloud data of the building thermal insulation material, and treat the sum of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material and the standard deviation of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material as the denoising distance of the point cloud data of the building thermal insulation material; The denoising unit is used to treat all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building insulation material, whose distance from the corresponding three-dimensional point is less than the denoising distance of the point cloud data of the building insulation material, as the denoising points of the point cloud data of the building insulation material.

[0039] The beneficial effects of the above technology are: a specific method for obtaining all denoised points of the point cloud data of building insulation materials based on all the neighboring three-dimensional points of each three-dimensional point of the point cloud data of building insulation materials is given in detail, thereby improving the quality and availability of the point cloud data.

[0040] Example 4: Based on Example 1, a building insulation material detection system based on laser point cloud, the processing module includes: a processing submodule, configured to use the position of the building insulation material when the three-dimensional laser scanner scans the building insulation material as the origin of the region division of the point cloud data of the building insulation material; The division submodule is used to take the area division origin of the point cloud data of the building thermal insulation material as the sphere center, 1 unit length and 0 unit length as the radius, obtain the spherical layer area (the spherical layer area that does not overlap between the sphere with the area division origin of the point cloud data of the building thermal insulation material as the sphere center, the sphere with 1 unit length as the radius, and the sphere with 0 unit length as the radius), and use the obtained spherical layer area as the first division area of the point cloud data of the building thermal insulation material, and regard all denoised points located in the first division area of the point cloud data of the building thermal insulation material as a denoised point set of the point cloud data of the building thermal insulation material, and take the area division origin of the point cloud data of the building thermal insulation material as the sphere center, the sphere with 2 unit lengths and 1 unit Length is used as the radius to obtain the spherical layer area (the spherical layer area that does not overlap between a sphere with a radius of 2 unit lengths and a sphere with a radius of 1 unit length, with the area division origin of the point cloud data of the building thermal insulation material as the sphere center), and the obtained spherical layer area is used as the second division area of the point cloud data of the building thermal insulation material. All denoised points located in the second division area of the point cloud data of the building thermal insulation material are regarded as a denoised point set of the point cloud data of the building thermal insulation material. The above-mentioned division area determination process is continued in a cycle (the process of continuing to determine the new spherical layer area) until all denoised points of the point cloud data of the building thermal insulation material have been delineated as denoised point sets, and all denoised point sets of the point cloud data of the building thermal insulation material are obtained.

[0041] In this embodiment, the three-dimensional laser scanner is a high-precision measurement device that quickly acquires three-dimensional coordinate data (X, Y, Z) of an object's surface by emitting a laser beam and measuring the laser reflection time or phase change.

[0042] The beneficial effects of the above technology are: all denoised points of the point cloud data of building insulation materials are divided into regions, and all denoised point sets of the point cloud data of building insulation materials are obtained, which facilitates the subsequent calculation of the cut-off value of each denoised point set of the point cloud data of building insulation materials.

[0043] Example 5: Based on Example 4, a building insulation material detection system based on laser point cloud, the computing module includes: A first evaluation submodule is configured to use the quotient of the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material and the volume of a divided area corresponding to the corresponding denoised point set (the corresponding divided area required to determine the denoised point set) as the density value of the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the quotient of the density value of each denoised point set of the point cloud data of the building thermal insulation material and the average of the density values of all denoised point sets of the point cloud data of the building thermal insulation material as the density evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; The second evaluation submodule is used to obtain an integrity evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The third evaluation submodule is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The calculation submodule is used to obtain the cut-off value of each denoised point set of the point cloud data of the building insulation material based on the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of the building insulation material.

[0044] In this embodiment, based on the density evaluation value, integrity evaluation value, and process evaluation value of each denoised point set of the point cloud data of the building insulation material, the cut-off value of each denoised point set of the point cloud data of the building insulation material is obtained, namely: ; Among them, α is the cut-off value of the denoised point set currently calculated for the point cloud data of building thermal insulation materials, A is the density evaluation value of the denoised point set currently calculated for the point cloud data of building thermal insulation materials, B is the integrity evaluation value of the denoised point set currently calculated for the point cloud data of building thermal insulation materials, and C is the process evaluation value of the denoised point set currently calculated for the point cloud data of building thermal insulation materials. is the mean value of the density evaluation value of all denoised point sets of the point cloud data of building insulation materials, is the mean of the integrity evaluation values of all denoised point sets of the point cloud data of building insulation materials, is the mean of the process evaluation values of all denoised point sets of the point cloud data of building insulation materials, is the minimum value of the density evaluation value of all denoised point sets of the point cloud data of building insulation materials, is the minimum value of the integrity evaluation value of all denoised point sets of the point cloud data of building insulation materials, is the minimum value of the process evaluation value of all denoised point sets of point cloud data of building insulation materials, is the standard deviation of the density evaluation value of all denoised point sets of the point cloud data of building insulation materials, is the standard deviation of the integrity evaluation value of all denoised point sets of the point cloud data of building insulation materials, is the standard deviation of the process evaluation value of all denoised point sets of the point cloud data of building insulation materials, n is the number of all denoised point sets of the point cloud data of building insulation materials, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0045] The beneficial effects of the above technology are: based on all denoised point sets of point cloud data of building insulation materials, the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of point cloud data of building insulation materials are obtained, and the distribution density, spatial distribution completeness and difficulty of the acquisition process of each denoised point set of point cloud data of building insulation materials are quantified separately. Then, based on the density evaluation value, integrity evaluation value and process evaluation value of all denoised point sets of point cloud data of building insulation materials, the cut-off value of each denoised point set of point cloud data of building insulation materials is obtained, and the priority of retaining each denoised point set of point cloud data of building insulation materials in the laser material detection process is accurately quantified.

[0046] Example 6: Based on Example 5, the second evaluation submodule of the building thermal insulation material detection system based on laser point cloud includes: an assignment unit for obtaining a volume of a cavity region of each denoised point set of the point cloud data of the building thermal insulation material based on each denoised point set of the point cloud data of the building thermal insulation material and a preset cavity detection model; A first evaluation unit is configured to use the reciprocal of a quotient between the volume of a hole region of each denoised point set of the point cloud data of the building thermal insulation material and the volume of a divided region corresponding to the denoised point set as an integrity evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; Among them, the preset void detection model is composed of a large number of denoised point sets and their corresponding divided areas of the pre-collected point cloud data of building insulation materials as model input, and the void area volume of the corresponding denoised point set manually marked and delineated as model output. The trained single denoised point set and its corresponding divided area that can be input into the point cloud data of building insulation materials can output the void area volume of the corresponding denoised point set of the point cloud data of building insulation materials.

[0047] In this embodiment, the volume of the hole region of the denoised point set is the volume occupied by the region not occupied by the denoised points (ie, the hole) in the divided region where all the denoised points in the denoised point set are located.

[0048] The beneficial effects of the above technology are: based on all denoised point sets of the point cloud data of building insulation materials, the integrity evaluation value of each denoised point set of the point cloud data of building insulation materials is obtained, and the degree of completeness of the spatial distribution of denoised points in each denoised point set of the point cloud data of building insulation materials is accurately quantified, which facilitates the calculation of the cut-off value of the subsequent denoised point set.

[0049] Example 7: Based on Example 5, the building thermal insulation material detection system based on laser point cloud, the third evaluation submodule includes: An acquisition unit, used for acquiring acquisition time of all denoised points in each denoised point set of point cloud data of building thermal insulation materials; The second evaluation unit is configured to use the average of the acquisition times of all denoised points in all denoised point sets of the point cloud data of the building thermal insulation material as the classification time value of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are less than the classification time value of the point cloud data of the building thermal insulation material as the first-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are not less than the classification time value of the point cloud data of the building thermal insulation material as the second-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material; The third evaluation unit is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all first-category denoised points and all second-category denoised points of each denoised point set of the point cloud data of the building thermal insulation material.

[0050] In this embodiment, the acquisition time of the denoised point is the length of time from the start of laser detection to the completion of data acquisition of the denoised point, for example, 1 second.

[0051] In this embodiment, based on all first-category denoised points and all second-category denoised points in each denoised point set of the point cloud data of the building thermal insulation material, a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material is obtained, namely: ; Among them, β is the process evaluation value of the current calculated denoised point set of the point cloud data of building insulation materials, The total number of all first-class denoised points in the currently calculated denoised point set for the point cloud data of building insulation materials, The total number of all second-category denoised points in the currently calculated denoised point set for the point cloud data of building insulation materials, is the numerical value of the classification time value of the point cloud data of building insulation materials, The sum of the acquisition times of all first-class denoised points in the currently calculated denoised point set of the point cloud data of the building insulation material is the value of the sum of the acquisition times of all first-class denoised points in the currently calculated denoised point set. It is the minimum value of the acquisition time of all the first-class denoised points in the currently calculated denoised point set of the point cloud data of the building insulation material. It is the maximum value of the acquisition time of all first-category denoised points in the currently calculated denoised point set of the point cloud data of the building insulation material, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0052] The beneficial effects of the above technology are: based on all denoised point sets of point cloud data of building insulation materials, the process evaluation value of each denoised point set of point cloud data of building insulation materials is obtained, and the difficulty of collecting denoised points in each denoised point set of point cloud data of building insulation materials is accurately quantified, which facilitates the calculation of the cut-off value of the subsequent denoised point set.

[0053] Example 8: Based on Example 1, a building insulation material detection system based on laser point cloud, the detection module includes: a rejection threshold determination submodule, for using the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material as the horizontal coordinate and the rejection value of the corresponding denoised point set as the vertical coordinate to obtain all rejection points of the point cloud data of the building thermal insulation material, and using each rejection point of the point cloud data of the building thermal insulation material as a circle point and a preset second distance as a radius to obtain a rejection circle area of each rejection point of the point cloud data of the building thermal insulation material, and using the number of remaining rejection points within the rejection circle area of each rejection point of the point cloud data of the building thermal insulation material as the rejection statistic value of the corresponding rejection point of the point cloud data of the building thermal insulation material, using the rejection point with the largest rejection statistic among all the rejection points of the point cloud data of the building thermal insulation material as the threshold point of the point cloud data of the building thermal insulation material, and using the vertical coordinate of the threshold point of the point cloud data of the building thermal insulation material as the rejection threshold of the point cloud data of the building thermal insulation material; The detection submodule obtains the detection results of the building insulation materials based on the cut-off threshold of the point cloud data of the building insulation materials and the cut-off value of all denoised point sets.

[0054] In this embodiment, the preset second distance is a preset distance value of a cut-off circle area for determining each cut-off point of the point cloud data of the building thermal insulation material.

[0055] In this embodiment, the cut-off threshold of the point cloud data of the building insulation material is used to compare with the cut-off value of each denoised point set of the point cloud data of the building insulation material to determine whether the corresponding denoised point set will be retained during the laser material detection process.

[0056] The beneficial effects of the above technology are: based on the cut-off values of all denoised point sets of the point cloud data of building insulation materials, the inspection results of building insulation materials are obtained. In terms of comprehensive inspection quality and inspection efficiency, the denoised points that need to be retained in the point cloud data of building insulation materials during the laser material inspection process are accurately screened out, greatly improving the inspection efficiency.

[0057] Example 9: Based on Example 8, the building insulation material detection system based on laser point cloud, the detection submodule includes: a first detection unit, configured to determine whether a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is greater than a cut-off threshold of the point cloud data of the building thermal insulation material; if so, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a discarded denoised point set of the point cloud data of the building thermal insulation material; The second detection unit is used to re-divide the divided area corresponding to each discarded denoised point set of the point cloud data of the building thermal insulation material, obtain all subsets of each discarded denoised point set of the point cloud data of the building thermal insulation material, and judge whether the cut-off value of each subset of each discarded denoised point set of the point cloud data of the building thermal insulation material is greater than (when calculating the cut-off value of each subset of each discarded denoised point set, each subset of each discarded denoised point set is regarded as a denoised point set, and if the discarded denoised point set is divided again, the discarded denoised point set is not regarded as a denoised point set when calculating the cut-off value, and the cut-off value calculation of a single subset of the discarded denoised point set of the newly obtained point cloud data of the building thermal insulation material does not affect the previously completed cut-off value calculation process, and there is no need to recalculate the previously completed cut-off value calculation process). If the rejection threshold of the point cloud data of the building thermal insulation material is met, then the corresponding subset of the discarded denoised point set of the point cloud data of the building thermal insulation material is used as the retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the corresponding subset of the discarded denoised point set of the point cloud data of the building thermal insulation material is used as the discarded denoised point set of the point cloud data of the building thermal insulation material, and the above-mentioned re-division process is continued to be repeated for the divided area corresponding to the discarded denoised point set of the newly obtained point cloud data of the building thermal insulation material until the volume of the divided area corresponding to the discarded denoised point set of the point cloud data of the building thermal insulation material obtained the most recently is less than the preset stop division volume, and then the re-division of the divided area corresponding to the discarded denoised point set of the point cloud data of the building thermal insulation material is stopped to obtain all the retained denoised point sets of the point cloud data of the building thermal insulation material; The third detection unit is used to obtain the contour detection result of the building thermal insulation material based on all denoised points of all retained denoised point sets of the point cloud data of the building thermal insulation material, and use the contour detection result of the building thermal insulation material as the detection result of the building thermal insulation material.

[0058] In this embodiment, the retained denoised point set is a set of denoised points that need to be retained during the laser material detection process.

[0059] In this embodiment, the discarded denoised point set is a set of denoised points that do not need to be retained during the laser material detection process.

[0060] In this embodiment, the divided area corresponding to the discarded denoised point set of the newly obtained point cloud data of the building insulation material is divided into two equal parts, and the two divided areas with the same volume obtained are both whole. In this embodiment, the bisection is performed by a pre-set division model (a large number of pre-collected division areas are used as model input, and the division results of the corresponding division areas divided by manual annotation are used as model output. The trained neural network model that can output a single division area can output the bisection result of the corresponding division area).

[0061] In this embodiment, all subsets of each discarded denoised point set of the point cloud data of the building insulation material are two subsets determined after re-dividing the divided area corresponding to each discarded denoised point set of the point cloud data of the building insulation material, and each subset corresponds to a re-divided divided area (all denoised points located in a single sub-divided area after the divided area corresponding to each discarded denoised point set is re-divided are regarded as a subset of the corresponding discarded denoised point set).

[0062] In this embodiment, the preset stop division volume is a preset volume value used to determine the stop of the re-division.

[0063] In this embodiment, the contour detection result of the building thermal insulation material is a detection and drawing result of the contour of the building thermal insulation material obtained by using all denoised points in all retained denoised point sets of the point cloud data of the building thermal insulation material.

[0064] The beneficial effects of the above technology are: based on the cut-off threshold of the point cloud data of the building insulation material and the cut-off value of all denoised point sets, the detection results of the building insulation material are obtained, and a specific method for screening out all denoised points of all retained denoised point sets of the point cloud data of the building insulation material is given in detail.

[0065] Example 10: The present invention provides a method for detecting building thermal insulation materials based on laser point cloud, which is applied to execute any one of the building thermal insulation material detection systems based on laser point cloud in Examples 1 to 9, with reference to Figure 2 ,include: S1: De-noising the collected point cloud data of the building thermal insulation material to obtain all the de-noised points of the point cloud data of the building thermal insulation material; S2: Divide all denoised points of the point cloud data of the building thermal insulation material into regions to obtain all denoised point sets of the point cloud data of the building thermal insulation material; S3: Based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material are obtained, and based on the density evaluation value, integrity evaluation value, and process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is obtained; S4: Obtaining a detection result of the building thermal insulation material based on the cut-off value of each denoised point set of the point cloud data of the building thermal insulation material.

[0066] The beneficial effects of the above technology are: all denoised points of the point cloud data of building thermal insulation materials are divided into regions, and all denoised point sets of the point cloud data of building thermal insulation materials are obtained, which is convenient for the subsequent calculation of the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials. According to all denoised point sets of the point cloud data of building thermal insulation materials, the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of building thermal insulation materials are obtained, and the distribution density, spatial distribution integrity and acquisition process difficulty of each denoised point set of the point cloud data of building thermal insulation materials are quantified separately, and then the density of the denoised points in each denoised point set of the point cloud data of building thermal insulation materials are quantified separately. The density evaluation value, integrity evaluation value and process evaluation value of all denoised point sets of the point cloud data of thermal insulation materials are calculated, and the cut-off value of each denoised point set of the point cloud data of building thermal insulation materials is obtained. This realizes the accurate quantification of the priority of retaining each denoised point set of the point cloud data of building thermal insulation materials in the process of laser material detection. Finally, according to the cut-off value of all denoised point sets of the point cloud data of building thermal insulation materials, the detection result of building thermal insulation materials is obtained. In terms of comprehensive detection quality and detection efficiency, the denoised points that need to be retained in the point cloud data of building thermal insulation materials in the process of laser material detection are accurately screened out, which greatly improves the detection efficiency.

[0067] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, and the present invention is also intended to include these changes and modifications.

Claims

1. A building insulation material detection system based on laser point cloud, characterized in that: include: An acquisition module is used to perform denoising on the collected point cloud data of the building thermal insulation material to obtain all denoised points of the point cloud data of the building thermal insulation material; a processing module, configured to divide all denoised points of the point cloud data of the building thermal insulation material into regions, and obtain all denoised point sets of the point cloud data of the building thermal insulation material; a calculation module for obtaining, based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, and obtaining, based on the density evaluation value, the integrity evaluation value, and the process evaluation value of all denoised point sets of the point cloud data of the building thermal insulation material, a cut-off value for each denoised point set of the point cloud data of the building thermal insulation material; The detection module is used to obtain the detection result of the building insulation material based on the cut-off value of all denoised point sets of the point cloud data of the building insulation material.

2. The building insulation material detection system based on laser point cloud according to claim 1 is characterized in that: Get modules, including: The acquisition submodule is used to obtain all three-dimensional points of the point cloud data of building insulation materials; a preprocessing submodule for obtaining a noise point detection space for each 3D point in the point cloud data of the building thermal insulation material, using each 3D point in the point cloud data of the building thermal insulation material as a sphere center and a preset first distance as a radius, and treating remaining 3D points within the noise point detection space for each 3D point in the point cloud data of the building thermal insulation material as neighboring 3D points of the corresponding 3D point in the point cloud data of the building thermal insulation material; The denoising submodule is used to obtain all denoised points of the point cloud data of the building thermal insulation material based on all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material.

3. The building insulation material detection system based on laser point cloud according to claim 2 is characterized in that: Denoising submodule, including: a preprocessing unit, configured to treat the distance between each three-dimensional point of the point cloud data of the building thermal insulation material and each neighboring three-dimensional point of the corresponding three-dimensional point as the neighboring distance of each neighboring three-dimensional point of each three-dimensional point of the point cloud data of the building thermal insulation material, treat the mean of the neighboring distances of all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building thermal insulation material as the mean of the neighboring distances of each three-dimensional point of the point cloud data of the building thermal insulation material, and treat the sum of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material and the standard deviation of the mean of the neighboring distances of all three-dimensional points of the point cloud data of the building thermal insulation material as the denoising distance of the point cloud data of the building thermal insulation material; The denoising unit is used to treat all neighboring three-dimensional points of each three-dimensional point of the point cloud data of the building insulation material, whose distance from the corresponding three-dimensional point is less than the denoising distance of the point cloud data of the building insulation material, as the denoising points of the point cloud data of the building insulation material.

4. The building insulation material detection system based on laser point cloud according to claim 1, characterized in that: Processing module, including: a processing submodule, configured to use the position of the building insulation material when the three-dimensional laser scanner scans the building insulation material as the origin of the region division of the point cloud data of the building insulation material; The division submodule is used to use the area division origin of the point cloud data of the building thermal insulation material as the center of the sphere, 1 unit length and 0 unit length as the radius, to obtain the spherical layer area, and use the obtained spherical layer area as the first division area of the point cloud data of the building thermal insulation material, and regard all denoised points located in the first division area of the point cloud data of the building thermal insulation material as a denoised point set of the point cloud data of the building thermal insulation material; use the area division origin of the point cloud data of the building thermal insulation material as the center of the sphere, 2 unit lengths and 1 unit length as the radius, to obtain the spherical layer area, and use the obtained spherical layer area as the second division area of the point cloud data of the building thermal insulation material, and regard all denoised points located in the second division area of the point cloud data of the building thermal insulation material as a denoised point set of the point cloud data of the building thermal insulation material; continue the cycle of the above-mentioned division area determination process until all denoised points of the point cloud data of the building thermal insulation material have been delineated as denoised point sets, and obtain all denoised point sets of the point cloud data of the building thermal insulation material.

5. The building insulation material detection system based on laser point cloud according to claim 4 is characterized in that: Computing module, including: A first evaluation submodule is configured to use the quotient of the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material and the volume of the divided area corresponding to the corresponding denoised point set as the density value of the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the quotient of the density value of each denoised point set of the point cloud data of the building thermal insulation material and the average of the density values of all denoised point sets of the point cloud data of the building thermal insulation material as the density evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; The second evaluation submodule is used to obtain an integrity evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The third evaluation submodule is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all denoised point sets of the point cloud data of the building thermal insulation material; The calculation submodule is used to obtain the cut-off value of each denoised point set of the point cloud data of the building insulation material based on the density evaluation value, integrity evaluation value and process evaluation value of each denoised point set of the point cloud data of the building insulation material.

6. The building insulation material detection system based on laser point cloud according to claim 5, characterized in that: The second evaluation submodule includes: an assignment unit for obtaining a volume of a cavity region of each denoised point set of the point cloud data of the building thermal insulation material based on each denoised point set of the point cloud data of the building thermal insulation material and a preset cavity detection model; A first evaluation unit is configured to use the reciprocal of a quotient between the volume of a hole region of each denoised point set of the point cloud data of the building thermal insulation material and the volume of a divided region corresponding to the denoised point set as an integrity evaluation value of the corresponding denoised point set of the point cloud data of the building thermal insulation material; Among them, the preset void detection model is composed of a large number of denoised point sets and their corresponding divided areas of the pre-collected point cloud data of building insulation materials as model input, and the void area volume of the corresponding denoised point set manually marked and delineated as model output. The trained single denoised point set and its corresponding divided area that can be input into the point cloud data of building insulation materials can output the void area volume of the corresponding denoised point set of the point cloud data of building insulation materials.

7. The building thermal insulation material detection system based on laser point cloud according to claim 5, characterized in that: The third evaluation submodule includes: An acquisition unit, used for acquiring acquisition time of all denoised points in each denoised point set of point cloud data of building thermal insulation materials; The second evaluation unit is configured to use the average of the acquisition times of all denoised points in all denoised point sets of the point cloud data of the building thermal insulation material as the classification time value of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are less than the classification time value of the point cloud data of the building thermal insulation material as the first-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material, and to use the denoised points in each denoised point set of the point cloud data of the building thermal insulation material whose acquisition times are not less than the classification time value of the point cloud data of the building thermal insulation material as the second-category denoised points in the corresponding denoised point set of the point cloud data of the building thermal insulation material; The third evaluation unit is used to obtain a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material based on all first-category denoised points and all second-category denoised points of each denoised point set of the point cloud data of the building thermal insulation material.

8. The building insulation material detection system based on laser point cloud according to claim 1, characterized in that: Detection module, including: a rejection threshold determination submodule, for using the number of denoised points in each denoised point set of the point cloud data of the building thermal insulation material as the horizontal coordinate and the rejection value of the corresponding denoised point set as the vertical coordinate to obtain all rejection points of the point cloud data of the building thermal insulation material, and using each rejection point of the point cloud data of the building thermal insulation material as a circle point and a preset second distance as a radius to obtain a rejection circle area of each rejection point of the point cloud data of the building thermal insulation material, and using the number of remaining rejection points within the rejection circle area of each rejection point of the point cloud data of the building thermal insulation material as the rejection statistic value of the corresponding rejection point of the point cloud data of the building thermal insulation material, using the rejection point with the largest rejection statistic among all the rejection points of the point cloud data of the building thermal insulation material as the threshold point of the point cloud data of the building thermal insulation material, and using the vertical coordinate of the threshold point of the point cloud data of the building thermal insulation material as the rejection threshold of the point cloud data of the building thermal insulation material; The detection submodule obtains the detection results of the building insulation materials based on the cut-off threshold of the point cloud data of the building insulation materials and the cut-off value of all denoised point sets.

9. The building insulation material detection system based on laser point cloud according to claim 8, characterized in that: Detection submodule, including: a first detection unit, configured to determine whether a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is greater than a cut-off threshold of the point cloud data of the building thermal insulation material; if so, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the corresponding denoised point set of the point cloud data of the building thermal insulation material is regarded as a discarded denoised point set of the point cloud data of the building thermal insulation material; The second detection unit is used to re-divide the divided area corresponding to each discarded denoised point set of the point cloud data of the building thermal insulation material, obtain all subsets of each discarded denoised point set of the point cloud data of the building thermal insulation material, and judge whether the cut-off value of each subset of each discarded denoised point set of the point cloud data of the building thermal insulation material is greater than the cut-off threshold of the point cloud data of the building thermal insulation material. If so, the corresponding subset of the discarded denoised point set of the point cloud data of the building thermal insulation material is regarded as the retained denoised point set of the point cloud data of the building thermal insulation material; otherwise, the point cloud data of the building thermal insulation material is retained. The corresponding subset of the corresponding discarded denoised point set of the cloud data is used as the discarded denoised point set of the point cloud data of the building thermal insulation material, and the above-mentioned re-division process is continued to be repeated for the divided area corresponding to the discarded denoised point set of the newly obtained point cloud data of the building thermal insulation material until the volume of the divided area corresponding to the discarded denoised point set of the point cloud data of the latest obtained building thermal insulation material is less than the preset stop division volume, and the re-division of the divided area corresponding to the discarded denoised point set of the point cloud data of the building thermal insulation material is stopped, and all the retained denoised point sets of the point cloud data of the building thermal insulation material are obtained; The third detection unit is used to obtain the contour detection result of the building thermal insulation material based on all denoised points of all retained denoised point sets of the point cloud data of the building thermal insulation material, and use the contour detection result of the building thermal insulation material as the detection result of the building thermal insulation material.

10. A method for detecting building thermal insulation materials based on laser point cloud, characterized in that: A building insulation material detection system based on laser point cloud, used for executing any one of claims 1 to 9, comprising: S1: De-noising the collected point cloud data of the building thermal insulation material to obtain all the de-noised points of the point cloud data of the building thermal insulation material; S2: Divide all denoised points of the point cloud data of the building thermal insulation material into regions to obtain all denoised point sets of the point cloud data of the building thermal insulation material; S3: Based on all denoised point sets of the point cloud data of the building thermal insulation material, a density evaluation value, an integrity evaluation value, and a process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material are obtained, and based on the density evaluation value, integrity evaluation value, and process evaluation value of each denoised point set of the point cloud data of the building thermal insulation material, a cut-off value of each denoised point set of the point cloud data of the building thermal insulation material is obtained; S4: Obtaining a detection result of the building thermal insulation material based on the cut-off value of each denoised point set of the point cloud data of the building thermal insulation material.

Images