A method and apparatus for monitoring a construction site

Abstract

The application provides a construction site monitoring method and device, the construction site comprises a tower crane, and a plurality of sensing assemblies are arranged on the tower crane, the sensing assemblies comprise a plurality of cameras and laser radars, the method comprises the following steps: receiving data collected by the sensing assemblies, the collected data comprises point cloud data collected by the laser radars and image data collected by the cameras; performing data fusion on the collected data to obtain fused data; and uploading the fused data to the cloud, so that a background can analyze the fused data to monitor the construction site. In the foregoing manner, the cost can be reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a construction site monitoring method and device. BACKGROUND

[0002] Tower crane is the most commonly used hoisting equipment in construction site, also known as "tower crane", which is used to hoist construction raw materials such as steel bars, wood blocks, concrete and steel pipes by connecting one section after another. Tower crane is an essential equipment in construction site.

[0003] When using tower crane, safe operation is a very important rule. During the operation of tower crane, the hook is far away from the cab, so it is difficult to monitor accurately. Therefore, a monitoring camera is usually installed above the hook to monitor the goods being transported or lifted, which is convenient for the workers to work.

[0004] The existing construction site monitoring and construction progress management technology mainly relies on two-dimensional picture information of monitoring camera, without three-dimensional information, and the fragmented information needs to be integrated before use.

[0005] The existing construction site scanning scheme uses unmanned aerial vehicle combined with laser radar for scanning. However, the cost of single scanning is very high, so most construction sites perform construction site scanning once a week or even a month to control the cost. SUMMARY

[0006] The present application provides a construction site monitoring method and device, which solves the technical problem of high scanning cost in the prior art.

[0007] The technical scheme of the present application is as follows: a construction site monitoring method is provided, the construction site includes a tower crane and a plurality of sensing components arranged on the tower crane, the sensing components include a plurality of cameras and a laser radar, the plurality of cameras are arranged at different positions of the tower crane, and the method comprises:

[0008] receiving data collected by the sensing components, the collected data including point cloud data collected by the laser radar and image data collected by the camera;

[0009] performing data fusion on the collected data to obtain fused data;

[0010] uploading the fused data to the cloud, so that the background can analyze the fused data to monitor the construction site.

[0011] In an optional manner, the tower crane is one, and the tower crane includes one sensing component. The collected data further includes inertial data. The data fusion on the collected data to obtain fused data comprises:

[0012] obtain preliminary state data based on the inertial data;

[0013] update the preliminary state data based on the adjacent two frames of image data;

[0014] perform denoising and coupling processing on the point cloud data based on the updated preliminary state data to obtain fusion data.

[0015] In an optional manner, the updating of the preliminary state data based on the adjacent two frames of image data comprises:

[0016] calculating transformation data of the adjacent two frames of image;

[0017] calculating a global pose of the current frame of image based on the point cloud data;

[0018] updating the preliminary state data based on the transformation data and the global pose.

[0019] In an optional manner, the cranes are a plurality of, and one of the cranes corresponds to one of the sensing assemblies. The data fusion on the collected data to obtain fusion data comprises:

[0020] obtaining a homogeneous matrix based on relative positions between the sensing assemblies to rotate and translate the point cloud data;

[0021] adjusting the overlapped point cloud data after the translation to obtain the fusion data.

[0022] In an optional manner, the adjusting of the overlapped point cloud data after the translation to obtain the fusion data comprises:

[0023] performing decomposition processing on each point cloud data in the overlapped region;

[0024] calculating distances between each point cloud point in the overlapped region and a point cloud where the point cloud point is based on a decomposition processing result;

[0025] summing up the calculated distances to obtain a corresponding error value, and obtaining an error value corresponding to each point cloud;

[0026] eliminating, based on the error value, a point cloud point in the overlapped region that does not satisfy a preset condition;

[0027] returning to the decomposition processing on the remaining point cloud data in the overlapped region until a change amount of the homogeneous matrix is less than a preset value.

[0028] In an optional manner, the calculation of the distances between each point cloud point in the overlapped region and the point cloud where the point cloud point is based on the decomposition processing result comprises:

[0029] Traverse each point cloud of the overlapping area, obtain the distance between each point cloud point and other point clouds, and obtain a distance set corresponding to each point cloud;

[0030] The calculated distances are summed to obtain corresponding error values, and the error values corresponding to each point cloud are obtained, including:

[0031] For each point cloud, the distances in the distance set of the point cloud are summed, and the corresponding distance sum is taken as an error value;

[0032] In an optional manner, the error values are used to eliminate point clouds in the overlapping area that do not meet a preset condition, including:

[0033] Minimizing the error values;

[0034] Eliminating point clouds that do not meet a preset condition.

[0035] The application further provides a monitoring device for a construction site, the construction site including a tower crane, and a sensing assembly arranged on the tower crane, the sensing assembly including a camera and a laser radar, and the device including:

[0036] A receiving module is configured to receive data collected by the sensing assembly, the collected data including point cloud data collected by the laser radar and image data collected by the camera;

[0037] A fusion module is configured to perform data fusion on the collected data to obtain fused data;

[0038] A transmission module is configured to upload the fused data to a cloud server, so that a background server can analyze the fused data to monitor the construction site.

[0039] The application further provides a computing device including a processor, a memory, a communication interface, and a communication bus, the processor, the memory, and the communication interface being in communication with each other via the communication bus;

[0040] The memory is configured to store at least one executable instruction, and the executable instruction causes the processor to perform steps of a monitoring method for the construction site.

[0041] The application further provides a computer storage medium, and the storage medium stores at least one executable instruction, and the executable instruction causes a processor to perform steps of a monitoring method for a construction site.

[0042] Compared with the prior art, the application has the following beneficial effects: the camera and the laser radar arranged on the tower crane are used to collect data, and the collected data is fused to obtain colorful three-dimensional image data, so that a background server can monitor the environment of the construction site, and the monitoring can be effectively performed, the cost can be reduced, and the efficiency can be improved. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments are briefly introduced below. The drawings described below are only the corresponding drawings of some embodiments of the present invention.

[0044] Figure 1 The diagram illustrates a flowchart of a construction site monitoring method provided by an embodiment of the present invention.

[0045] Figure 2 This diagram illustrates the structure of a monitoring device for a construction site according to an embodiment of the present invention.

[0046] Figure 3 The diagram shows a schematic of the sensing component structure of a monitoring device for a construction site provided in an embodiment of the present invention;

[0047] Figure 4 The diagram shows a partial schematic of a monitoring system provided by an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention is shown. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] The directional terms mentioned in this invention, such as "up", "down", "front", "back", "left", "right", "inner", "outer", "side", "top" and "bottom", are only for reference to the orientation of the accompanying drawings. The directional terms used are for the purpose of explaining and understanding this invention, and are not intended to limit this invention.

[0051] The terms "first" and "second" used in the terminology of this invention are for descriptive purposes only and should not be construed as indicating or implying relative importance, nor as limiting the order of events.

[0052] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting", "fixing" and the like should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, it can be the internal communication of two elements or the interaction relationship of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0053] Please refer to Figure 1 The preferred embodiment of the present application provides a flowchart of a construction site monitoring method. In the embodiment of the present application, the monitoring method is used to monitor a construction site, the construction site includes at least one tower crane, the number of tower cranes is determined according to actual needs, which is not limited here, further, the tower crane is provided with a plurality of sensing components, the sensing components include a plurality of cameras and laser radars, the camera can be a camera, the number of cameras can be set according to actual conditions, which is not limited here. The number of laser radars is determined according to the number of tower cranes, and in general, one laser radar is configured for one tower crane to collect point cloud data, the method comprises:

[0054] Step S1, receiving data collected by the sensing component;

[0055] Specifically, the sensing component can collect data of the environment of the construction site in real time, or collect data periodically. In order to reduce the cost, it is preferred to collect data periodically, and the collection period can be set according to actual conditions, which is not limited here. The laser radar collects point cloud data, and the camera collects image data.

[0056] Step S2, data fusion is performed on the collected data to obtain fusion data;

[0057] Specifically, the collected point cloud data and image data are fused to obtain fusion data.

[0058] Step S3, uploading the fusion data to the cloud, so that the background can analyze the fusion data to monitor the construction site;

[0059] Specifically, the fusion data is uploaded to the cloud, so that the staff in the background can analyze the environment of the construction site according to the fusion data to monitor reasonably.

[0060] In the embodiment of the present application, the camera and laser radar arranged on the tower crane are used to collect data, and the collected data is fused to obtain colorful three-dimensional image data, which is convenient for the background to monitor the environment of the construction site, can effectively monitor, reduce the cost and improve the efficiency.

[0061] In an optional mode of the embodiment, the time of data collection by the camera and the laser radar is synchronized, and both have the same collection period.

[0062] In an optional mode of the embodiment, the crane tower is one, and the crane tower includes the sensor assembly, and the collected data further includes inertial data, and the step S2 includes:

[0063] The preliminary state data is obtained based on the inertial data;

[0064] Specifically, the sensor assembly includes the camera and the laser radar, and further includes an inertial measurement unit (IMU) inside the laser radar, wherein the sensor assembly is controlled by a pulse signal generated by a single-chip microcomputer, the feedback speed of the inertial measurement unit is fast, the feedback speed of the camera is medium, and the feedback speed of the laser radar is slow, the inertial data of the position of the crane tower is measured by the inertial measurement unit, the state data of the position of the crane tower at the current time point is calculated by integral recursion and based on Newton's second law, and the state data can include the current position and the corresponding speed, and can further include other data such as the azimuth angle, which is not limited here;

[0065] The preliminary state data is updated based on the adjacent two frames of image data.

[0066] Specifically, for the multiple frames of image collected by the camera, the adjacent two frames of image are compared, and the preliminary state data is updated according to the comparison result.

[0067] The point cloud data is denoised and coupled based on the updated preliminary state data to obtain fusion data.

[0068] Specifically, the preliminary state data is updated, the current global map and local map are updated, the point cloud in the current visible range of the camera is projected to the image coordinate system, the current color value and the color data recorded in the map can be obtained, the color value and the color data are modeled based on the foregoing color value and color data, the final fusion data is obtained through Gaussian noise coupling processing, preferably, Kalman filtering is used to calculate the gain and update the color value of each point cloud, that is, each point cloud is colored, and finally the fusion data is obtained.

[0069] In a further preferred embodiment of the embodiment, the process of updating the preliminary state data based on the adjacent two frames of image data is as follows:

[0070] The transformation data of the adjacent two frames of image is calculated.

[0071] Specifically, the transformation data between the two frames of image is calculated based on an optical flow tracking algorithm, wherein the optical flow tracking algorithm uses existing technology, which is not described here.

[0072] calculating a global pose of the current frame image based on the point cloud data;

[0073] Specifically, based on the point cloud data of the current position, the corresponding point cloud is projected to the current frame to calculate the global pose of the current frame image, wherein the existing technology can be used to calculate the global pose.

[0074] updating the preliminary state data based on the transformation data and the global pose.

[0075] Specifically, based on the transformation data and the global pose, the preliminary state data is updated by minimizing the photometric error of the current frame image to obtain the fusion data.

[0076] In another embodiment of the present application, the hanger towers are multiple, and one of the hanger towers corresponds to one of the sensing assemblies, and the step S2 specifically comprises:

[0077] Based on the relative positions between the sensing assemblies, a homogeneous matrix is obtained to rotate and translate the point cloud data.

[0078] Specifically, since the sensing assemblies are arranged on the hanger towers at different positions, a homogeneous matrix is obtained according to the relative positions of the sensing assemblies to rotate and translate the point cloud, wherein the point cloud data includes color data of each point cloud and coordinate data of each point cloud. At this time, the coordinate data of the point cloud is preferably rotated and translated.

[0079] The overlapping point cloud data after translation is adjusted to obtain the fusion data.

[0080] Specifically, after the aforementioned point cloud is rotated and translated, some point clouds will overlap, and at this time, the point clouds in the overlapping region need to be adjusted to obtain the fusion data.

[0081] In one preferred mode of the present embodiment, the specific process of adjusting the overlapping point cloud data after translation to obtain the fusion data is as follows:

[0082] Each point cloud data in the overlapping region is decomposed.

[0083] Specifically, the point cloud data includes a plurality of point cloud points, the point cloud data of the overlapping region is decomposed, the point cloud points are divided into small blocks, and then the point cloud points in the small block region are analyzed to determine whether the point cloud points belong to a surface point cloud or a line point cloud, so as to obtain the surface point cloud or the line point cloud corresponding to each point. For example, the point cloud is a collection of a plurality of point cloud points, and the arrangement mode formed by the spatial distribution of the plurality of point cloud points is called a point cloud. The line point cloud is a line composed of a plurality of point cloud points, and the surface point cloud is a surface composed of a plurality of point cloud points. The aforementioned "small blocks" refer to a small range around each point cloud point, which is used to determine whether the point cloud point is on a line or a surface, on which line or surface.

[0084] Based on the decomposition processing result, the distance between each point in the overlapping region and the point cloud in which the point is located is calculated.

[0085] Specifically, the distance between the point (point cloud point) and the line point cloud or the surface point cloud in which the point is located is calculated according to the decomposed point type (surface point cloud or line point cloud), so as to obtain the distance between the point and the line or the surface. At this time, if the points in the overlapping region are relatively many, there will be multiple distance values. The distance between two points method, the point-to-line distance method, and the point-to-surface distance method can be used to calculate the corresponding distance. Preferably, each point cloud point in the overlapping region is traversed to obtain the distance between each point cloud point and the line or the surface (i.e., the surface point cloud or the line point cloud) formed by each point cloud in which the point cloud point is located, so as to obtain the distance between each point cloud point and the point cloud in which the point cloud point is located. The distances of the point cloud points obtained for each point cloud form a set, so as to obtain the distance set corresponding to each point cloud. It should be noted that each line point cloud and surface point cloud are fitted with the parameters of the line or the surface by the least square method. Since the existing algorithm is used here, the technical details of the least square method will not be described again.

[0086] The calculated distances are summed to obtain the corresponding error value, and the error value corresponding to each point cloud is obtained.

[0087] Specifically, the distances calculated above are summed to obtain the sum of the distances, and the sum of the distances is the error value of the point cloud, so as to obtain the error value of each point cloud in the overlapping region. For example, for each point cloud, the distances in the distance set of the point cloud are summed, and the sum of the distances is taken as the error value.

[0088] Based on the error value, the point cloud points in the overlapping region that do not meet the preset condition are removed.

[0089] Specifically, the error value is first minimized, wherein a gradient descent method is used to minimize the error value. The point cloud points that do not satisfy a preset condition are removed, such as comparing the error value of the point cloud in the overlapping area with a preset threshold value, and if the error value is greater than the preset threshold value, it is considered that the error comparison is large and needs to be removed, so the corresponding point cloud point in the overlapping area is removed. The preset condition refers to that the error value of the point cloud is less than or equal to the preset threshold value.

[0090] Then, the step of decomposing the remaining point cloud data in the overlapping area is turned to the aforementioned step until the change amount of the homogeneous matrix is less than a preset value.

[0091] Specifically, after removing part of the points in the overlapping area, the error value of the remaining points in the area needs to be re-calculated until the change amount of the homogeneous matrix is less than a preset value. The process is consistent with the above description and can be referred to the above description.

[0092] In the embodiment of the present application, the camera and the laser radar arranged on the tower crane are used to collect data, and the collected data is fused to obtain three-dimensional image data, which is convenient for the background to monitor the environment of the construction site, so as to effectively monitor and reduce the cost and improve the efficiency.

[0093] Secondly, the data collected at different time points can be compared to understand the progress of the construction project, and problems can be found and handled in time.

[0094] Furthermore, the positions of the workers can be identified for personnel management, and different equipment such as excavators and bulldozers can be identified for positioning and management.

[0095] Based on the above embodiment, the present application further provides a monitoring device for a construction site, as shown in the accompanying drawings: Figure 2 The construction site includes a tower crane and a sensing assembly arranged on the tower crane, and the sensing assembly includes a camera and a laser radar. The monitoring device 2 includes a receiving module 21, a fusion module 22 connected with the receiving module 21, and a transmission module 23 connected with the fusion module 22, wherein:

[0096] The receiving module 21 is used to receive the data collected by the sensing assembly, and the collected data includes point cloud data collected by the laser radar and image data collected by the camera.

[0097] The fusion module 22 is used to fuse the collected data to obtain fused data.

[0098] The transmission module 23 is used to upload the fused data to the cloud, so that the background can analyze the fused data to monitor the construction site.

[0099] In an optional manner, the fusion module 22 is specifically configured to:

[0100] respectively pre-process the image data and the point cloud data;

[0101] extract a pixel value of each pixel point from the pre-processed image data;

[0102] assign the pixel value of the pixel point to a corresponding point cloud point in the pre-processed point cloud data, so as to obtain a corresponding global image, wherein the coordinates of the point cloud point are equal to those of the corresponding pixel point.

[0103] In an optional manner, the cranes are one, and one of the cranes includes one of the sensing assemblies. The collected data further includes inertial data, and the fusion module 22 is specifically configured to:

[0104] obtain preliminary state data based on the inertial data;

[0105] update the preliminary state data based on adjacent two frames of image data;

[0106] perform denoising and coupling processing on the point cloud data based on the updated preliminary state data, so as to obtain fusion data.

[0107] In an optional manner, the fusion module 22 is specifically configured to:

[0108] calculate transformation data of adjacent two frames of image data;

[0109] calculate a global pose of a current frame of image data based on the point cloud data;

[0110] update the preliminary state data based on the transformation data and the global pose.

[0111] The cranes are multiple, and one of the cranes corresponds to one of the sensing assemblies. The fusion module 22 is specifically configured to:

[0112] obtain a homogeneous matrix based on relative positions between the sensing assemblies, and perform rotation and translation on the point cloud data;

[0113] perform adjustment processing on the translated and overlapped point cloud data, so as to obtain fusion data.

[0114] In an optional manner, the fusion module 22 is specifically configured to:

[0115] perform decomposition processing on each point cloud data in the overlapped region;

[0116] calculate distances between each point cloud point in the overlapped region and a point cloud in which the point cloud point is located based on a decomposition processing result;

[0117] sum the calculated distances to obtain a corresponding error value, so as to obtain an error value corresponding to each of the point clouds.

[0118] eliminating, based on the error value, a point cloud point in the overlapping area that does not satisfy a preset condition;

[0119] Turning to a step of decomposing the point cloud data remaining in the overlapping area until a change of the homogeneous matrix is less than a preset value.

[0120] In an optional manner, the fusion module 22 is specifically configured to:

[0121] traversing each point cloud point in the overlapping area, obtaining a distance between each point cloud point and a point cloud in which the point cloud point is located, and obtaining a distance set corresponding to each point cloud;

[0122] The fusion module 22 is specifically configured to:

[0123] for each point cloud, summing distances in the distance set of the point cloud, and taking a corresponding distance sum as an error value;

[0124] In an optional manner, the fusion module 22 is specifically configured to:

[0125] minimizing the error value;

[0126] eliminating a point cloud point that does not satisfy a preset condition.

[0127] In the embodiment of the present application, data is collected by a camera and a laser radar arranged on a tower crane, and the collected data is fused to obtain three-dimensional image data, which is convenient for monitoring the environment of a construction site in the background, so as to effectively monitor and reduce cost and improve efficiency.

[0128] Based on the above embodiment, the present application further provides a monitoring system, as shown in Figure 3 and Figure 4 The construction site is provided with at least one tower crane 31, and the monitoring system comprises a sensing assembly 32, the sensing assembly 32 comprising a camera 321, a laser radar 322 and a housing 323, wherein the camera 321 is arranged on the tower crane 31, the tower crane 31 is provided with the laser radar 322, and the monitoring device 2 described in the above embodiment is further included, the specific structure, working principle and technical effects of the monitoring device 2 are consistent with the description of the above embodiment, and details are not repeated here. In the embodiment, the monitoring device further comprises a housing 34, and the monitoring system further comprises a fixing mechanism 35, wherein the fixing mechanism 35 is fixedly connected with the housing 34, the fixing mechanism 35 is fixedly connected with a maintenance frame (not shown in the figure) of the tower crane, and the sensing assembly 32 is fixed on a trolley (not shown in the figure) of the tower crane 31.

[0129] In another preferred embodiment of the present application, the monitoring system further comprises a gimbal, which can replace the aforementioned fixed mechanism, i.e., the gimbal has the function of the fixed mechanism and can provide a more open view for the user.

[0130] In one preferred embodiment of the present application, the monitoring system further comprises a cloud platform connected with the monitoring device 2, which is used for background analysis and evaluation of data and timely monitoring of the construction site.

[0131] In one preferred embodiment of the present application, the crane tower 34 is multiple, and each crane tower 34 is provided with a camera 321 and a laser radar 322.

[0132] In the embodiment of the present application, the camera and the laser radar arranged on the crane tower are used to collect data, and the collected data is fused to obtain color three-dimensional image data, which is convenient for the background to monitor the environment of the construction site, so as to effectively monitor and reduce the cost and improve the efficiency.

[0133] The embodiment of the present application provides a non-volatile computer storage medium, which stores at least one executable instruction, and the computer executable instruction can execute the construction site monitoring method in any method embodiment.

[0134] The data collected by the sensing assembly includes point cloud data collected by the laser radar and image data collected by the camera;

[0135] The collected data is fused to obtain fused data;

[0136] The fused data is uploaded to the cloud, and the background analyzes the fused data to monitor the construction site.

[0137] In an optional manner, the crane tower is one, and the crane tower comprises the sensing assembly, and the collected data further comprises inertial data, and the executable instruction causes the processor to perform the following operations:

[0138] Preliminary state data is obtained based on the inertial data;

[0139] The preliminary state data is updated based on adjacent two frames of image data;

[0140] The point cloud data is denoised and coupled based on the updated preliminary state data to obtain fused data.

[0141] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0142] The transformation data of adjacent two frames of image is calculated;

[0143] calculating a global pose of the current frame image based on the point cloud data;

[0144] updating the preliminary state data based on the transformation data and the global pose.

[0145] In an alternative way, the crane towers are multiple, one of the crane towers corresponds to one of the sensing assemblies, and the executable instructions cause the processor to perform the following operations:

[0146] obtaining a homogeneous matrix based on the relative positions between the sensing assemblies, and rotating and translating the point cloud data;

[0147] adjusting the point cloud data in the overlapping region after the translation to obtain fusion data.

[0148] In an alternative way, the executable instructions cause the processor to perform the following operations:

[0149] decomposing each point cloud data in the overlapping region;

[0150] calculating the distance between each point cloud point in the overlapping region and the point cloud where the point cloud point is located based on the decomposition result;

[0151] summing the calculated distances to obtain a corresponding error value, and obtaining the error value corresponding to each point cloud;

[0152] eliminating the point cloud points in the overlapping region that do not satisfy a preset condition based on the error value;

[0153] returning to the step of decomposing the point cloud data in the remaining overlapping region until the change of the homogeneous matrix is less than a preset value.

[0154] In an alternative way, the executable instructions cause the processor to perform the following operations:

[0155] traversing each point cloud point in the overlapping region, obtaining the distance between each point cloud point and the point cloud where the point cloud point is located, and obtaining a distance set corresponding to each point cloud;

[0156] The executable instructions cause the processor to perform the following operations:

[0157] for each point cloud, summing the distances in the distance set of the point cloud, and taking the corresponding distance sum as an error value;

[0158] In an alternative way, the executable instructions cause the processor to perform the following operations:

[0159] minimizing the error value;

[0160] eliminating the point cloud points that do not satisfy a preset condition.

[0161] In the embodiment of the present application, the data is collected by the camera and the laser radar arranged on the tower, and the collected data is fused to obtain color three-dimensional image data, so that the environment of the construction site can be monitored in the background, and the monitoring efficiency is improved.

[0162] Figure 5 The structural schematic diagram of the computing device provided by the embodiment of the present application is shown, and the embodiment of the present application does not limit the specific implementation of the device.

[0163] As shown in the figure, the computing device can include a processor 502, a communications interface 504, a memory 506, and a communications bus 508. Figure 5

[0164] The processor 502, the communications interface 504, and the memory 506 can communicate with each other through the communications bus 508. The communications interface 504 is configured to communicate with network elements such as clients or other servers. The processor 502 is configured to execute the program 510, and specifically can execute the related steps in the three-dimensional imaging method of the inverter.

[0165] Specifically, the program 510 can include program code including computer operation instructions.

[0166] The processor 502 can be a central processing unit CPU, or an application specific integrated circuit ASIC, or one or each integrated circuit configured to implement the embodiment of the present application. One or each processor included in the device can be the same type of processor, such as one or each CPU; or can be different types of processors, such as one or each CPU and one or each ASIC.

[0167] The memory 506 is configured to store the program 510. The memory 506 can include a high-speed RAM memory, and can also include a non-volatile memory such as at least one disk memory.

[0168] The program 510 can specifically be used to make the processor 502 perform the following operations:

[0169] Receive the data collected by the sensing component, and the collected data includes point cloud data collected by the laser radar and image data collected by the camera;

[0170] ​Data fusion is performed on the collected data to obtain fused data.

[0171] The fused data is uploaded to the cloud, and the background analyzes the fused data to monitor the construction site.

[0172] In an optional manner, the crane tower is one, and the crane tower includes the sensing assembly. The collected data further includes inertial data. The program 510 can be specifically configured to cause the processor 502 to perform the following operations:

[0173] Preliminary state data is obtained based on the inertial data.

[0174] The preliminary state data is updated based on adjacent two frames of image data.

[0175] The point cloud data is denoised and coupled based on the updated preliminary state data to obtain fused data.

[0176] In an optional manner, the program 510 can be specifically configured to cause the processor 502 to perform the following operations:

[0177] Transformation data of adjacent two frames of images is calculated.

[0178] Global poses of the current frame of images are calculated based on the point cloud data.

[0179] The preliminary state data is updated based on the transformation data and the global poses.

[0180] In an optional manner, the crane tower is multiple, and one crane tower corresponds to one sensing assembly. The program 510 can be specifically configured to cause the processor 502 to perform the following operations:

[0181] A rotation matrix is obtained based on relative positions between the sensing assemblies, and the point cloud data is rotated and translated.

[0182] The fused data is obtained by adjusting the translated and overlapped point cloud data.

[0183] In an optional manner, the program 510 can be specifically configured to cause the processor 502 to perform the following operations:

[0184] Each point cloud data in the overlapping region is decomposed.

[0185] Distances between each point cloud point in the overlapping region and the point cloud where the point cloud point is located are calculated based on the decomposition result.

[0186] The calculated distances are summed to obtain corresponding error values, and the error values corresponding to each point cloud are obtained.

[0187] Eliminate the point cloud points in the overlapping area that do not satisfy the preset condition based on the error value.

[0188] Turning to the step of decomposing the point cloud data remaining in the overlapping area until the change of the homogeneous matrix is less than a preset value.

[0189] In an optional manner, the program 510 can be specifically used for causing the processor 502 to perform the following operations:

[0190] Traverse each point cloud point in the overlapping area, obtain the distance between each point cloud point and the point cloud where the point cloud point is located, and obtain the distance set corresponding to each point cloud;

[0191] The program 510 can be specifically used for causing the processor 502 to perform the following operations:

[0192] For each point cloud, sum the distances in the distance set of the point cloud, and take the corresponding distance sum as an error value;

[0193] In an optional manner, the program 510 can be specifically used for causing the processor 502 to perform the following operations:

[0194] Minimize the error value;

[0195] Eliminate the point cloud points that do not satisfy the preset condition.

[0196] In the embodiment of the application, the data is collected by the camera and the laser radar arranged on the tower, and the collected data is fused to obtain colorful three-dimensional image data, so that the environment of the construction site can be monitored in the background, and the monitoring efficiency can be improved.

[0197] To sum up, although the application has been disclosed as above with preferred embodiments, the above preferred embodiments are not used to limit the application, and those skilled in the art can make various changes and decorations without departing from the spirit and scope of the application, therefore, the protection scope of the application is subject to the range defined by the claims.

Claims

1. A method of monitoring a construction site, characterized by, The construction site comprises a tower crane, and a sensing assembly arranged on the tower crane, the sensing assembly comprising a camera and a laser radar, the method comprising: receiving data collected by the sensing assembly, the collected data comprising point cloud data collected by the laser radar and image data collected by the camera; performing data fusion on the collected data to obtain fused data; wherein, if the tower crane is multiple, one sensing assembly corresponds to one tower crane, each sensing assembly is arranged on a tower crane at a different position, and the performing data fusion on the collected data to obtain fused data comprises: obtaining a homogeneous matrix based on the relative positions between each sensing assembly, and rotating and translating the point cloud data; adjusting the overlapped point cloud data after translation to obtain fused data; uploading the fused data to the cloud, so as to facilitate the background to analyze the fused data for monitoring the construction site; wherein, the adjusting the overlapped point cloud data after translation to obtain fused data comprises the steps of: performing decomposition processing on each point cloud data in the overlapping region; calculating the distance between each point cloud point in the overlapping region and the point cloud and line cloud where it is located based on the decomposition processing result; summing the calculated distances to obtain a corresponding error value, to obtain an error value corresponding to each point cloud; eliminating point cloud points in the overlapping region that do not meet a preset condition based on the error value; returning to the step of performing decomposition processing on the remaining point cloud data in the overlapping region until the change amount of the homogeneous matrix is less than a preset value.

2. The monitoring method according to claim 1, characterized in that, The tower crane is one, one sensing assembly comprises a camera and a laser radar, the method comprising: receiving data collected by the sensing assembly, the collected data comprising point cloud data collected by the laser radar and image data collected by the camera; performing data fusion on the collected data to obtain fused data; wherein, if the tower crane is multiple, one sensing assembly corresponds to one tower crane, each sensing assembly is arranged on a tower crane at a different position, and the performing data fusion on the collected data to obtain fused data comprises: obtaining a homogeneous matrix based on the relative positions between each sensing assembly, and rotating and translating the point cloud data; 3. The monitoring method according to claim 2, characterized in that, adjusting the overlapped point cloud data after translation to obtain fused data; uploading the fused data to the cloud, so as to facilitate the background to analyze the fused data for monitoring the construction site; wherein, the adjusting the overlapped point cloud data after translation to obtain fused data comprises the steps of: performing decomposition processing on each point cloud data in the overlapping region; 4. The monitoring method according to claim 3, characterized in that, calculating the distance between each point cloud point in the overlapping region and the point cloud and line cloud where it is located based on the decomposition processing result; summing the calculated distances to obtain a corresponding error value, to obtain an error value corresponding to each point cloud; eliminating point cloud points in the overlapping region that do not meet a preset condition based on the error value; returning to the step of performing decomposition processing on the remaining point cloud data in the overlapping region until the change amount of the homogeneous matrix is less than a preset value.

5. The monitoring method according to claim 4, characterized in that, The tower crane is one, one sensing assembly comprises a camera and a laser radar, the method comprising: receiving data collected by the sensing assembly, the collected data comprising point cloud data collected by the laser radar and image data collected by the camera; performing data fusion on the collected data to obtain fused data; 6. A construction site monitoring device, characterized by wherein, if the tower crane is multiple, one sensing assembly corresponds to one tower crane, each sensing assembly is arranged on a tower crane at a different position, and the performing data fusion on the collected data to obtain fused data comprises: obtaining a homogeneous matrix based on the relative positions between each sensing assembly, and rotating and translating the point cloud data; adjusting the overlapped point cloud data after translation to obtain fused data; uploading the fused data to the cloud, so as to facilitate the background to analyze the fused data for monitoring the construction site; wherein, the adjusting the overlapped point cloud data after translation to obtain fused data comprises the steps of: performing decomposition processing on each point cloud data in the overlapping region; calculating the distance between each point cloud point in the overlapping region and the point cloud and line cloud where it is located based on the decomposition processing result; summing the calculated distances to obtain a corresponding error value, to obtain an error value corresponding to each point cloud; eliminating point cloud points in the overlapping region that do not meet a preset condition based on the error value; returning to the step of performing decomposition processing on the remaining point cloud data in the overlapping region until the change amount of the homogeneous matrix is less than a preset value. The tower crane is one, one sensing assembly comprises a camera and a laser radar, the method comprising: receiving data collected by the sensing assembly, the collected data comprising point cloud data collected by the laser radar and image data collected by the camera; performing data fusion on the collected data to obtain fused data; wherein, if the tower crane is multiple, one sensing assembly corresponds to one tower crane, each sensing assembly is arranged on a tower crane at a different position, and the performing data fusion on the collected data to obtain fused data comprises: obtaining a homogeneous matrix based on the relative positions between each sensing assembly, and rotating and translating the point cloud data; adjusting the overlapped point cloud data after translation to obtain fused data; uploading the fused data to the cloud, so as to facilitate the background to analyze the fused data for monitoring the construction site; wherein, the adjusting the overlapped point cloud data after translation to obtain fused data comprises the steps of: performing decomposition processing on each point cloud data in the overlapping region; calculating the distance between each point cloud point in the overlapping region and the point cloud and line cloud where it is located based on the decomposition processing result; summing the calculated distances to obtain a corresponding error value, to obtain an error value corresponding to each point cloud; eliminating point cloud points in the overlapping region that do not meet a preset condition based on the error value; returning to the step of performing decomposition processing on the remaining point cloud data in the overlapping region until the change amount of the homogeneous matrix is less than a preset value. The receiving module is configured to receive data collected by the sensing assembly, wherein the collected data comprises point cloud data collected by the laser radar and image data collected by the camera. The fusion module is configured to perform data fusion on the collected data to obtain fused data. If the cranes are multiple, one of the cranes corresponds to one of the sensing assemblies, and each of the sensing assemblies is arranged on a different crane. The data fusion on the collected data to obtain the fused data comprises obtaining a homogeneous matrix based on relative positions between the sensing assemblies, and performing rotation and translation on the point cloud data. The transmission module is configured to upload the fused data to a cloud server, so that a background server can analyze the fused data to monitor the construction site. The adjustment processing on the translated point cloud data in the overlapping region to obtain the fused data comprises the following steps: decomposing each point cloud data in the overlapping region; calculating distances between each point cloud point in the overlapping region and a face point cloud and a line point cloud where the point cloud point is located based on a decomposition result; summing the calculated distances to obtain an error value, so as to obtain an error value corresponding to each point cloud; eliminating, based on the error value, a point cloud point in the overlapping region that does not satisfy a preset condition; returning to the decomposition processing on the remaining point cloud data in the overlapping region until a change amount of the homogeneous matrix is less than a preset value.

7. A computing device, comprising: The system comprises: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction, and the executable instruction causes the processor to execute steps of the monitoring method of the construction site according to any one of claims 1-5.

8. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction, and the executable instruction causes the processor to execute steps of the monitoring method of the construction site according to any one of claims 1-5.

Images