A method and system for quickly constructing an infrared three-dimensional model

Abstract

The present invention relates to the technical field of image data processing, and particularly relates to a method and system for quickly constructing an infrared three-dimensional model. The method includes: obtaining point cloud data of a target scene; dividing the point cloud data into multiple blocks from different perspectives; determining the degree of noise influence of each block among the multiple blocks; performing noise reduction processing on the point cloud data based on the degree of noise influence to obtain denoised data; and constructing an infrared three-dimensional model of the target scene based on the denoised data. This method can reduce the interference and noise caused to the point cloud data; moreover, by using the denoised data to perform a three-dimensional model of the target scene, the constructed three-dimensional model can be made more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a method and system for quickly constructing an infrared three-dimensional model in the technical field of image data processing. Background Art

[0002] An infrared three-dimensional model generally refers to a three-dimensional model generated by processing data obtained using infrared technology through a computer. This technology is mainly used in thermal imaging and infrared photography and can be used to analyze information such as the thermal distribution and temperature changes of an object. By collecting infrared image data of a target or scene and performing preprocessing, and using computer vision and image processing technologies, the three-dimensional structure and shape of an object are inferred from the preprocessed infrared images.

[0003] In related technologies, in order to construct an infrared three-dimensional model of a target ship, the target object is scanned with infrared light. During the scanning process, the collected point cloud data will be interfered and thus generate noise due to transmission and environmental impacts. Currently, the common method for noise filtering of point cloud data is to directly use existing filtering algorithms to perform noise filtering on the entire point cloud data. However, due to the different reflection capabilities of objects to infrared light, the noise sensitivities of different parts are different. Therefore, when performing filtering operations, it is necessary to analyze different parts for regional filtering. When directly using existing filtering algorithms to perform noise filtering on point cloud data, it is difficult to achieve the purpose of regional filtering of different parts of an object. Summary of the Invention

[0004] In order to solve the technical problem that it is difficult to achieve regional filtering of different parts of an object when using existing filtering algorithms to perform noise filtering on point cloud data, the purpose of the present invention is to provide a method and system for quickly constructing an infrared three-dimensional model. The specific technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for quickly constructing an infrared three-dimensional model. The method includes:

[0006] Step 101, obtaining point cloud data of a target scene;

[0007] Step 102, dividing the point cloud data from different perspectives to obtain multiple blocks;

[0008] Step 103, determining the noise influence degree of each block in the multiple blocks;

[0009] Step 104, performing noise reduction processing on the point cloud data based on the noise influence degree to obtain denoised data;

[0010] Step 105, constructing an infrared three-dimensional model of the target scene based on the denoised data.

[0011] Preferably, the specific steps of step 101 include:

[0012] Use an infrared distance sensor to collect data on the target scene to obtain multiple distance data;

[0013] Preprocess the multiple distance data to obtain the point cloud data.

[0014] Preferably, the specific steps of step 102 include:

[0015] Cluster the point cloud data from different perspectives to obtain multiple clustering clusters;

[0016] Segment different regions corresponding to the multiple clustering clusters to obtain the multiple blocks.

[0017] Preferably, segmenting different regions corresponding to the multiple clustering clusters to obtain the multiple blocks includes:

[0018] For any one of the multiple clustering clusters, determine the projection plane of the any one clustering cluster in the clustering cluster space coordinate system;

[0019] Among multiple data points in the projection plane, determine the neighborhood of each data point;

[0020] Based on the neighborhoods of the respective data points, divide different regions corresponding to the multiple clustering clusters into the multiple blocks.

[0021] Preferably, based on the neighborhoods of the respective data points, dividing different regions corresponding to the multiple clustering clusters into the multiple blocks includes:

[0022] Based on the maximum value points in the neighborhood, determine the smoothness of any one data point;

[0023] Determine the data density within the neighborhood of any one data point;

[0024] Based on the smoothness and the data density, determine the loss value for region segmentation within any one clustering cluster;

[0025] Based on the loss value, divide different regions corresponding to the multiple clustering clusters into the multiple blocks.

[0026] Preferably, based on the loss value, dividing different regions corresponding to the multiple clustering clusters into the multiple blocks includes:

[0027] Among the loss values corresponding to each clustering cluster, determine the minimum loss value;

[0028] Divide different regions in the clustering cluster corresponding to the minimum loss value into the multiple blocks.

[0029] Preferably, the specific steps of step 103 include:

[0030] In any one of the multiple sub-blocks, determine the minimum point;

[0031] In any one of the sub-blocks, determine the candidate minimum points adjacent to the minimum point, and the maximum points outside the preset radius of the minimum point;

[0032] Based on the minimum point, the candidate minimum points and the maximum points, determine the gully coefficient of any one of the sub-blocks;

[0033] Based on the gully coefficient of any one of the sub-blocks, determine the noise influence degree of any one of the sub-blocks.

[0034] Preferably, based on the minimum point, the candidate minimum points and the maximum points, determining the gully coefficient of any one of the sub-blocks includes:

[0035] In any one of the sub-blocks, determine the first distance between the minimum point and the candidate minimum points, and the second distance between the minimum point and the maximum points;

[0036] Determine the difference between the minimum point and the maximum point;

[0037] Based on the difference, the first distance and the second distance, determine the gully coefficient of any one of the sub-blocks.

[0038] Preferably, based on the gully coefficient of any one of the sub-blocks, determining the noise influence degree of any one of the sub-blocks includes:

[0039] In any one of the sub-blocks, determine the abnormal data points;

[0040] Based on the mean value of the abnormal data points and the gully coefficient, determine the noise influence degree of any one of the sub-blocks.

[0041] Preferably, the specific steps of step 104 include:

[0042] Use the noise influence degree as the proportional factor of Wiener filtering to filter the point cloud data to obtain the denoised data.

[0043] Second, a fast construction system for an infrared three-dimensional model is provided, and the system includes:

[0044] An acquisition module for acquiring the point cloud data of the target scene;

[0045] A division module for dividing the point cloud data from different perspectives to obtain a plurality of sub-blocks;

[0046] A determination module, configured to determine the noise influence degree of each block among the multiple blocks;

[0047] A noise reduction module, configured to perform noise reduction processing on the point cloud data based on the noise influence degree to obtain denoised data;

[0048] A construction module, configured to construct an infrared three-dimensional model of the target scene based on the denoised data.

[0049] In a third aspect, a computer program product is provided, which includes computer program code that, when running on a computer, causes the computer to execute the method in the first aspect or any one of the possible implementation manners described in the first aspect.

[0050] In a fourth aspect, a computer-readable storage medium is provided, which stores computer program code that, when running on a computer, causes the computer to execute the method in the first aspect or any one of the possible implementation manners described in the first aspect.

[0051] Compared with the prior art, a method for quickly constructing an infrared three-dimensional model of the present invention has the beneficial effects that: after obtaining the point cloud data of the infrared image of the target scene, the infrared image is divided into multiple blocks from different perspectives through the point cloud data, so that the image data from different perspectives in the infrared image can be reflected by the multiple blocks. Then, the noise influence degree of each block among the multiple blocks is determined. In this way, the noise situation in each block can be accurately characterized by the noise influence degree, which is convenient for subsequent noise reduction processing of the point cloud data based on the noise influence degree to obtain denoised data; finally, a three-dimensional model of the target scene is constructed according to the denoised data. In this way, when performing a filtering operation, by analyzing the noise influence degree of each block from different perspectives, it is convenient to perform filtering on different parts, so as to reduce the interference and noise caused to the point cloud data; moreover, by using the denoised data to perform a three-dimensional model of the target scene, the constructed three-dimensional model can be made more accurate. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without creative efforts.

[0053] Figure 1It is a schematic diagram of the implementation process of a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention;

[0054] Figure 2 It is a schematic diagram of the application scenario of a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention;

[0055] Figure 3 It is another schematic diagram of the implementation process of a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention;

[0056] Figure 4 It is still another schematic diagram of the implementation process of a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention;

[0057] Figure 5 It is a schematic diagram of the composition structure of a system for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention;

[0058] Figure 6 It is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0059] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of a method for quickly constructing an infrared three-dimensional model proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0060] The following specifically describes the specific solution of a method for quickly constructing an infrared three-dimensional model provided by the present invention with reference to the accompanying drawings.

[0061] Please refer to Figure 1 , which shows a flowchart of the implementation of a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention. The method can be implemented through the following steps:

[0062] Step 101, obtain the point cloud data of the target scene.

[0063] Here, the data of the target scene is collected by using an infrared distance sensor, and a plurality of distances are obtained. Then, the plurality of distances are preprocessed to obtain the point cloud data. For example, the data of the target scene is collected from multiple directions by using an infrared distance sensor, and the obtained plurality of distances are normalized to obtain the point cloud data. In this way, by collecting infrared data from multiple directions of the target scene, the obtained point cloud data is richer.

[0064] In some possible implementations, taking the ship shown in the target scenario as an example, for a ship that requires a 3D model to be constructed, since the construction of the 3D model involves the spatial positions of various parts of the ship, it is necessary to collect data from all aspects of the ship. Figure 2 For the ship, select a suitable UAV and install an infrared distance sensor suitable for the ship above the UAV. Start the UAV and turn on the infrared distance sensor, and scan from above the ship to obtain the distance data of the scene. Among them, the distance data includes: the distance information of multiple points on the ship. Thus, the distance data of the ship is obtained, and the distance data is normalized to obtain point cloud data. Point cloud is a data set composed of a large number of discrete points. The position of each point is determined by the measurement value of the ranging sensor, and the data of each point is the 3D point cloud data.

[0065] Step 102: Divide the point cloud data into multiple blocks from different perspectives.

[0066] Here, the point cloud data is divided according to the multiple perspectives included in the target scenario, so as to obtain multiple blocks. Among them, different blocks are used to represent different regions in the target scenario.

[0067] In some possible implementations, for the point cloud data of the target scenario obtained by the above operations, since the subsequent construction of the 3D model is based on the point cloud data, but due to environmental factors and other reasons during infrared ranging, noise will appear in the obtained point cloud data, and it is necessary to denoise the obtained point cloud data. Therefore, it is necessary to project the discrete data points of the obtained point cloud data onto different planes. Analyze the data points projected onto different planes to obtain different blocks among them; according to the noise sensitivity of the blocks, obtain the degree of noise influence of each block, and perform denoising processing on different blocks respectively according to the degree of noise influence. The above step 102 can be implemented through

[0068] the steps shown in Figure 3 the following:

[0069] Step 301: Cluster the point cloud data from different perspectives to obtain multiple clustering clusters.

[0070] Here, taking the target scenario of a naval vessel as an example, for the point cloud data obtained from the above operations, since noise analysis is to be performed on the collected point cloud data of the naval vessel and the naval vessels from different perspectives need to be analyzed, it is necessary to cluster the discrete three-dimensional data points in the point cloud data and the internal situation of the scattered points from each perspective can be better reflected within each clustering cluster. Therefore, the data points in the clustering cluster should be able to reflect the characteristics of each surface of the naval vessel, and the difference between the data after the clustering cluster is projected should be small; otherwise, it is a point cluster that does not belong to this surface. Therefore, several data point clusters of the surfaces of the ship are obtained through K-means clustering, that is, multiple clustering clusters are obtained.

[0071] Step 302, segment different regions corresponding to the multiple clustering clusters to obtain the multiple blocks.

[0072] Here, taking the target scenario of a naval vessel as an example, after obtaining the data point clustering clusters of different surfaces of the naval vessel, since there are many different partitions within one surface of the naval vessel, such as the deck, cockpit, etc., there are different partitions on different surfaces, and due to the surface materials of the partitions and the complexity of the surface flatness under the perspective of this clustering cluster being different. When performing infrared ranging, due to the differences between the materials and the flatness, there are differences in the density and flatness between the obtained point cluster data, and thus the division of different regions can be obtained, thereby obtaining multiple blocks.

[0073] In some possible implementation manners, the above step 302 can be implemented through the following steps 321 to 323:

[0074] Step 321, for any one of the multiple clustering clusters, determine the projection plane of the any one clustering cluster in the clustering cluster space coordinate system.

[0075] For example, for any one clustering cluster, project the current observation azimuth onto the projection plane perpendicular to the observation azimuth, and the projection plane is the x and y axes in the clustering cluster space coordinate system, so as to obtain several new clustering point clusters with different perspective reference systems.

[0076] Step 322, among the multiple data points in the projection plane, determine the neighborhood of each data point.

[0077] For example, for the data points in each projection plane, determine the surrounding area with a radius of 3 centered on it, so as to obtain the neighborhood of each data point.

[0078] Step 323, based on the neighborhoods of the data points, divide the different regions corresponding to the multiple clustering clusters into the multiple blocks.

[0079] Here, in the neighborhood of each data point, the data density of the neighborhood and the smoothness of each data point are calculated, so as to realize the division of different regions in the clustering cluster by combining the smoothness and the data density.

[0080] In some possible implementation manners, the above step 323 may be implemented through the following process:

[0081] First, based on the maximum points in the neighborhood, the smoothness of any data point is determined.

[0082] Here, for each clustering cluster, the scattered points therein are projected onto the xy plane in its new coordinate system, and the z-axis data is recorded as the value of the scattered point. The density is reflected by the discreteness of the scattered points on the projection plane, and the flatness is reflected by the values of the scattered points. For each data point on the projection plane, a surrounding area with a radius of 3 centered on it is determined, and the maximum points (if the value of a data point is larger in a circle with a radius of 1 centered on it, then it is recorded as a maximum point) and minimum points therein are obtained, and the data points between the maximum points are obtained. Thus, the smoothness of each data point in each clustering cluster can be obtained , as shown in formula (1):

[0083] (1);

[0084] Wherein, is any data point in the clustering cluster, is any maximum point in the surrounding area (i.e., neighborhood) of the data point in the clustering cluster, J represents the number of maximum points, is the maximum point closest to any maximum point in the surrounding area of the data point in the clustering cluster, is the th value of the th maximum point in the surrounding area of the th data point in the clustering cluster, is the th value of the maximum point closest to the th maximum point in the surrounding area of the th data point in the clustering cluster, is the exponential function with the natural constant

[0085] Secondly, determine the data density within the neighborhood of any one of the data points.

[0086] Here, the data density within the neighborhood of a data point is obtained by the number of data points in the area around the data point.

[0087] Thirdly, based on the smoothness degree and the data density, determine the loss value for region segmentation within any one of the clustering clusters.

[0088] Here, according to the smoothness degree and density of the data points, different neighborhoods are segmented for each different surface clustering cluster, and the loss value within the cluster after segmentation needs to be minimized. Among them, the loss value of the segmentation is E, as shown in formula (2):

[0089] (2);

[0090] Wherein, is any one neighborhood in the clustering cluster, N represents the number of neighborhoods, is the loss value of the segmentation within the clustering cluster, is the variance of the smoothness degree and density of the data points within the -th neighborhood in the clustering cluster, is the information entropy of the product of the mean values of the smoothness degree and density of the data points within each neighborhood in the clustering cluster, and this information entropy is not 0; is an exponential function with the natural constant e as the base.

[0091] Finally, based on the loss value, divide the different regions corresponding to the multiple clustering clusters into the multiple blocks. In this way, by combining the smoothness degree and the data density, the loss value of the region segmentation within each clustering cluster can be accurately calculated, so as to accurately divide the different regions corresponding to the clustering clusters into multiple blocks.

[0092] In some possible implementation manners, by determining the minimum loss value among the loss values corresponding to each clustering cluster; for example, after obtaining several new clustering point clusters with different perspective reference systems, calculate their loss values, where the loss value is the mean value of the maximum differences of the z - axis values of each clustering cluster in the new coordinate system. According to the obtained loss value, obtain the clustering cluster with the minimum loss value, because when the loss value is smaller, it indicates that the difference in the z - axis values of the data in the clustering point cluster is smaller, so it indicates that it is on the same surface. And divide the different regions in the clustering cluster with the minimum loss value into the multiple blocks. In this way, by calculating the loss corresponding to each clustering cluster and segmenting the different regions in the clustering cluster with the minimum loss value, the point cloud data can be accurately divided into multiple blocks, so as to facilitate subsequent filtering operations according to each block.

[0093] Step 103, determine the degree of noise influence of each block among the multiple blocks.

[0094] In some possible implementations, taking a warship as an example of the target scenario, for the obtained several clustering clusters and their new coordinate systems, since one clustering cluster reflects the scattered points distributed on one surface of the warship, and there are different blocks on different surfaces of the warship, such as the deck, the hull, the cockpit, the cargo, etc.; therefore, for the clustering clusters of each surface obtained, it is necessary to divide each block according to the differences in structural features within different blocks, and obtain the noise influence degree of each block according to the internal abnormal data points of each block and the sensitivity to the noise within each block. The above step 103 can be achieved through Figure 4 the steps shown as follows:

[0095] Step 401, in any one of the multiple blocks, determine the minimum value point.

[0096] Here, for different blocks in the clustering clusters obtained by the above operations, since different blocks belong to the same part of the warship; and since different parts of the warship surface have different accuracies in measuring due to different reflections and absorptions of infrared light during infrared ranging, resulting in different sensitivities to noise for different parts. If an abnormal data point appears in a part with high noise sensitivity, then the possibility that it belongs to noise is high, and vice versa; therefore, it is necessary to quantify the noise sensitivity according to the characteristics of the internal data points of different parts; among them, the gully part will cause some deviations due to the refraction of infrared light, so its noise sensitivity is relatively low. For any one block, obtain the values of each data point within the block, and take the value of the smallest data point as the minimum value point within the block.

[0097] Step 402, in the any one block, determine the candidate minimum value points adjacent to the minimum value point, and, the maximum value points outside the preset radius of the minimum value point.

[0098] Step 403, based on the minimum value point, the candidate minimum value points and the maximum value points, determine the gully coefficient of the any one block.

[0099] In some possible implementations, in the any one block, first, determine the first distance between the minimum value point and the candidate minimum value points, and, the second distance between the minimum value point and the maximum value points; secondly, determine the difference between the minimum value point and the maximum value points; finally, based on the difference, the first distance and the second distance, determine the gully coefficient of the any one block. In this way, by analyzing the difference between the maximum value point and the minimum value point, and the difference between the minimum value point and the surrounding candidate minimum value points, the gully coefficient of each block can be accurately analyzed.

[0100] Here, the gully coefficient of each sub-block is obtained according to the distribution of data points within the sub-block. Since the gully-shaped part will cause partial refraction effects during infrared light ranging and thus abnormal points are more likely to appear; the data values on both sides of the gully part are higher while the values inside are lower. Therefore, for the data points within the sub-block, the minimum value points and the maximum value points are determined (a data point is the maximum or minimum value within a circle with a radius of 1 centered on itself, denoted as the maximum value point or the minimum value point); the gully coefficient of the sub-block is obtained as shown in formula (3):

[0101] (3);

[0102] Among them, is any minimum value point in the clustering cluster sub-block, is the number of minimum value points in the clustering cluster sub-block, is the nearest minimum value point (i.e., the candidate minimum value point) around any minimum value point in the clustering cluster sub-block, is the nearest maximum value point outside the radius of 1 of any minimum value point in the clustering cluster sub-block, is the th gully coefficient of the th sub-block in the clustering cluster, is the th distance between the th minimum value point and its nearest minimum value point in the th sub-block in the clustering cluster, is the average value of the data points between the th minimum value point and its nearest minimum value point in the th sub-block in the clustering cluster, is the value of the th minimum value point in the th sub-block in the clustering cluster, is the value of the nearest minimum value point to the th minimum value point in the th sub-block in the clustering cluster, is the distance between the th minimum value point and the nearest maximum value point outside the radius of 1 of the th minimum value point in the th sub-block in the clustering cluster, is the difference between the th minimum value point and its nearest maximum value point outside the radius of 1 in the

[0103] For the first distance between the minimum value point and its nearest minimum value point (i.e., the candidate minimum value point) in the classification block of the clustering cluster , the larger its value, the smaller the first distance between the minimum points it is located at, thus indicating that it is more likely to appear in a gully shape; and for the interpolation of the numerical mean of the minimum point and its nearest minimum point and the numerical mean of the data points therebetween , the smaller its value, the smaller the numerical difference of the data points between the minimum points, thus indicating that it is easy to form a gully shape; and for the second distance between the extreme point and its nearest maximum point in the clustering cluster block , the smaller its value, the smaller the second distance between the minimum point and the maximum point in the image, thus indicating that it is more likely to appear in a gully shape; and for the numerical difference between the minimum point and the maximum point (that is, the difference between the minimum point and the maximum point), the larger its value, the deeper the gully shape, and thus the higher its gully coefficient.

[0104] Step 404, based on the gully coefficient of any block, determine the noise influence degree of any block.

[0105] In the above steps 401 to 404, for any block, by combining the maximum point and the minimum point in the block, first calculate the gully coefficient of any block, and then accurately calculate the noise influence degree of the block through the gully coefficient of the block.

[0106] In some possible implementation manners, by determining abnormal data points in any block; and based on the mean value of the abnormal data points and the gully coefficient, determine the noise influence degree of any block. In this way, by analyzing the abnormal data points in each block and combining the mean value of the abnormal data points with the gully coefficient of the block, the noise influence degree of the block can be calculated quickly and accurately.

[0107] Here, according to the obtained gully coefficient, obtain the noise influence degree of each block. First, obtain the abnormal points of each block according to the LOF algorithm. For the LOF mean value of the abnormal points, compare it with the gully coefficient of the block to obtain the noise influence degree, as shown in formula (4):

[0108] (4);

[0109] Among them, is the noise influence degree of the th block in the clustering cluster, is the LOF mean value of the abnormal points in the th block in the clustering cluster, is the hyperbolic function; represents the absolute value of.

[0110] Step 104: Based on the degree of noise influence, perform noise reduction processing on the point cloud data to obtain denoised data.

[0111] Here, according to the degree of noise influence of each block in the clustering clusters obtained from the above operations, Wiener filtering with better ability to retain original information is selected, and the degree of noise influence is used as the scaling factor of Wiener filtering; among them, the scaling factor is used to adjust the filtering intensity of Wiener filtering. Generally, the image signal-to-noise ratio is used to set the scaling factor, but the image signal-to-noise ratio cannot be obtained in the case of no clear reference image. Therefore, the information difference is used as the scaling factor of Wiener filtering, which can make Wiener filtering have the highest efficiency and the best filtering effect when filtering the noise in the blocks of each clustering cluster. In this way, using the degree of noise influence as the scaling factor of Wiener filtering, filter the point cloud data to achieve noise reduction processing, so as to obtain the denoised data after filtering is completed, that is, the point cloud data after filtering is completed.

[0112] Step 105: Based on the denoised data, construct a three-dimensional model of the target scene.

[0113] Here, convert the denoised data into a voxel grid. A voxel is a cubic unit in three-dimensional space. By filling the denoised data into the voxels, voxelized three-dimensional data can be created. Based on the method of Poisson equation, a smooth surface can be reconstructed from the disordered point cloud data; map the texture information to the surface of the generated three-dimensional model to increase the realism and details.

[0114] In the embodiment of the present invention, the infrared image is divided into multiple blocks from different perspectives through the point cloud data, so that the image data from different perspectives in the infrared image can be reflected by the multiple blocks. Then, determine the degree of noise influence of each block in the multiple blocks. In this way, the degree of noise influence can accurately characterize the noise situation in each block, which is convenient for subsequent noise reduction processing of the point cloud data based on the degree of noise influence to obtain denoised data; finally, construct a three-dimensional model of the target scene according to the denoised data. In this way, when performing filtering operations, by analyzing the degree of noise influence of each block from different perspectives, it is convenient to perform filtering on different parts, so as to reduce the interference and noise caused to the point cloud data; and through the denoised data, constructing a three-dimensional model of the target scene can make the constructed three-dimensional model more accurate.

[0115] The embodiment of the present invention provides a system for quickly constructing an infrared three-dimensional model, as Figure 5 shown, the system 500 includes:

[0116] An acquisition module 501, configured to acquire point cloud data of a target scene;

[0117] A partitioning module 502, configured to partition the point cloud data from different perspectives to obtain multiple chunks;

[0118] A determining module 503, configured to determine the degree of noise influence of each chunk among the multiple chunks;

[0119] A noise reduction module 504, configured to perform noise reduction processing on the point cloud data based on the degree of noise influence to obtain denoised data;

[0120] A construction module 505, configured to construct an infrared three-dimensional model of the target scene based on the denoised data.

[0121] In some possible implementation manners, the partitioning module 502 is further configured to perform clustering on the point cloud data from different perspectives to obtain multiple clustering clusters; segment different regions corresponding to the multiple clustering clusters to obtain the multiple chunks.

[0122] In some possible implementation manners, the partitioning module 502 is further configured to, for any one of the multiple clustering clusters, determine a projection plane of the any one clustering cluster in a clustering cluster space coordinate system; determine neighborhoods of each data point among multiple data points in the projection plane; and divide different regions corresponding to the multiple clustering clusters into the multiple chunks based on the neighborhoods of each data point.

[0123] In some possible implementation manners, the partitioning module 502 is further configured to determine the smoothness degree of any one data point based on maximum value points in the neighborhood; determine the data density within the neighborhood of the any one data point; determine a loss value for region segmentation within the any one clustering cluster based on the smoothness degree and the data density; and divide different regions corresponding to the multiple clustering clusters into the multiple chunks based on the loss value.

[0124] In some possible implementation manners, the partitioning module 502 is further configured to determine a minimum loss value among loss values corresponding to each clustering cluster; and divide different regions in the clustering cluster corresponding to the minimum loss value into the multiple chunks.

[0125] In some possible implementation manners, the determining module 503 is further configured to determine minimum value points in any one of the multiple chunks; determine candidate minimum value points adjacent to the minimum value points, and maximum value points outside a preset radius of the minimum value points in the any one chunk; determine a gully coefficient of the any one chunk based on the minimum value points, the candidate minimum value points, and the maximum value points; and determine the degree of noise influence of the any one chunk based on the gully coefficient of the any one chunk.

[0126] In some possible implementation manners, the determining module 503 is further configured to determine a first distance between a minimum point and a candidate minimum point, and a second distance between the minimum point and a maximum point in any one of the sub-blocks; determine a difference between the minimum point and the maximum point; and determine a gully coefficient of any one of the sub-blocks based on the difference, the first distance, and the second distance.

[0127] In some possible implementation manners, the determining module 503 is further configured to determine abnormal data points in any one of the sub-blocks; and determine the degree of noise influence of any one of the sub-blocks based on the mean value of the abnormal data points and the gully coefficient.

[0128] In some possible implementation manners, the obtaining module 501 is further configured to collect data of the target scene by using an infrared distance sensor to obtain a plurality of distance data; and preprocess the plurality of distance data to obtain the point cloud data.

[0129] Optionally, the transmission medium may be a wired link (such as, but not limited to, coaxial cable, optical fiber, and Digital Subscriber Line (DSL), etc.) or a wireless link (such as, but not limited to, Wireless Fidelity (WIFI), Bluetooth, and mobile device network, etc.).

[0130] It should be noted that: the devices provided in the above embodiments are only illustrated by dividing the above functional modules. In actual applications, the above functions may be allocated to different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0131] Figure 6 is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Exemplarily, as Figure 6 shown, the computer device 600 includes: a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602. When the processor 602 executes the computer program 603, the computer device can execute any one of the above-introduced methods for quickly constructing an infrared three-dimensional model.

[0132] In addition, an embodiment of the present invention further protects a device, which may include a memory and a processor. The memory stores executable program code, and the processor is configured to call and execute the executable program code to execute a method for quickly constructing an infrared three-dimensional model provided by an embodiment of the present invention.

[0133] In this embodiment, the device can be divided into functional modules according to the above method examples. For example, each functional module can be corresponding, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there can be other division methods in actual implementation. It should be noted that all relevant contents of each step involved in the above method embodiment can be cited in the function description of the corresponding functional module, and will not be elaborated here.

[0134] It should be understood that the device provided in this embodiment is used to execute the above method for quickly constructing an infrared three-dimensional model, so the same effect as the above implementation method can be achieved. In the case of using an integrated unit, the device can include a processing module and a storage module. Among them, when the device is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc. Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logical blocks, modules, and circuits described in combination with the disclosure of the present invention. The processor can also be a combination that realizes computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0135] In addition, the device provided in the embodiment of the present invention can specifically be a chip, a component, or a module. The chip can include a processor and a memory connected thereto; among them, the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the method for quickly constructing an infrared three-dimensional model provided in the above embodiment. This embodiment also provides a computer-readable storage medium, in which computer program code is stored. When the computer program code runs on a computer, the computer is enabled to execute the above relevant method steps to implement the method for quickly constructing an infrared three-dimensional model provided in the above embodiment.

[0136] It should be noted that: the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous. Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. The above content is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for rapidly constructing an infrared three-dimensional model, characterized in that: The method comprises: Obtain point cloud data of the target scene; Dividing the point cloud data from different perspectives to obtain multiple blocks; Determining a noise impact degree of each of the multiple blocks; Based on the noise impact degree, performing noise reduction processing on the point cloud data to obtain noise-reduced data; Based on the de-noised data, construct an infrared three-dimensional model of the target scene; The point cloud data is divided into multiple blocks according to different perspectives, including: Clustering the point cloud data from different perspectives to obtain multiple clusters; Segmenting different regions corresponding to the multiple clusters to obtain the multiple blocks; The segmenting of different regions corresponding to the multiple clusters to obtain the multiple blocks includes: For any cluster among the multiple clusters, determining a projection plane of the any cluster in the cluster space coordinate system; Determine a neighborhood of each data point among a plurality of data points in the projection plane; Based on the neighborhood of each data point, different areas corresponding to the multiple clusters are divided into the multiple blocks; The dividing different areas corresponding to the multiple clusters into the multiple blocks based on the neighborhood of each data point includes: Determining the smoothness of any data point based on the maximum point in the neighborhood; Determining the data density in the neighborhood of any one of the data points; Based on the smoothness and the data density, determining a loss value for performing region segmentation within any cluster; Based on the loss value, different regions corresponding to the multiple clusters are divided into the multiple blocks; The determining the noise influence degree of each block in the plurality of blocks comprises: In any one of the plurality of blocks, determining a minimum point; In any of the blocks, determine candidate minimum points adjacent to the minimum point, and maximum points outside a preset radius of the minimum point; Determine the gully coefficient of any of the blocks based on the minimum point, the candidate minimum point and the maximum point; Based on the gully coefficient of any of the sub-blocks, the noise influence degree of any of the sub-blocks is determined.

2. The method for rapidly constructing an infrared three-dimensional model according to claim 1, characterized in that: The dividing the different regions corresponding to the multiple clusters into the multiple blocks based on the loss value includes: Determine the minimum loss value among the corresponding loss values ​​in each cluster; Different regions in the cluster corresponding to the minimum loss value are divided into the multiple blocks.

3. The method for rapidly constructing an infrared three-dimensional model according to claim 1, characterized in that: The determining of the gully coefficient of any of the blocks based on the minimum point, the candidate minimum point and the maximum point comprises: In any of the blocks, determining a first distance between a minimum point and a candidate minimum point, and a second distance between the minimum point and the maximum point; Determine the difference between the minimum point and the maximum point; Based on the difference, the first distance and the second distance, a gully coefficient of any one of the blocks is determined.

4. The method for rapidly constructing an infrared three-dimensional model according to claim 1, characterized in that: The determining the noise influence degree of any block based on the gully coefficient of any block includes: In any of the blocks, determining abnormal data points; The noise influence degree of any one of the blocks is determined based on the mean value of the abnormal data points and the gully coefficient.

5. The method for rapidly constructing an infrared three-dimensional model according to claim 1, characterized in that: The step of obtaining the point cloud data of the target scene includes: Using an infrared distance sensor to collect data on the target scene to obtain a plurality of distance data; The plurality of distance data are preprocessed to obtain the point cloud data.

6. A rapid construction system for infrared three-dimensional models, characterized in that: The system comprises: An acquisition module is used to acquire point cloud data of the target scene; A division module, used for dividing the point cloud data from different perspectives to obtain multiple blocks; A determination module, used to determine the noise impact degree of each block in the multiple blocks; A noise reduction module, used for performing noise reduction processing on the point cloud data based on the noise influence degree to obtain noise-reduced data; A construction module, used for constructing an infrared three-dimensional model of the target scene based on the noise-reduced data; The point cloud data is divided into multiple blocks according to different perspectives, including: Clustering the point cloud data from different perspectives to obtain multiple clusters; Segmenting different regions corresponding to the multiple clusters to obtain the multiple blocks; The segmenting of different regions corresponding to the multiple clusters to obtain the multiple blocks includes: For any cluster among the multiple clusters, determining a projection plane of the any cluster in the cluster space coordinate system; Determine a neighborhood of each data point among a plurality of data points in the projection plane; Based on the neighborhood of each data point, different areas corresponding to the multiple clusters are divided into the multiple blocks; The dividing different areas corresponding to the multiple clusters into the multiple blocks based on the neighborhood of each data point includes: Determining the smoothness of any data point based on the maximum point in the neighborhood; Determining the data density in the neighborhood of any one of the data points; Based on the smoothness and the data density, determining a loss value for performing region segmentation within any cluster; Based on the loss value, different regions corresponding to the multiple clusters are divided into the multiple blocks; The determining the noise influence degree of each block in the plurality of blocks comprises: In any one of the plurality of blocks, determining a minimum point; In any of the blocks, determine candidate minimum points adjacent to the minimum point, and maximum points outside a preset radius of the minimum point; Determine the gully coefficient of any of the blocks based on the minimum point, the candidate minimum point and the maximum point; Based on the gully coefficient of any of the sub-blocks, the noise influence degree of any of the sub-blocks is determined.

Images